fix: restore labels, codicon aliases, and issue branch flow

.gitignore

@@ -16,3 +16,4 @@ artifacts/vsix/
 artifacts/release-notes/
 artifacts/**/*.log
 artifacts/perf/runtime-benchmarks.json
+artifacts/perf/*.json

README.md

@@ -104,7 +104,7 @@ Both `defaultHighlight` and `customHighlight` allow for the following settings:
 
 `borderRadius` - used to set the border radius of the background of the highlight.
 
-`icon` - used to set a different icon in the tree view. Must be a valid octicon (see <https://octicons.github.com>) or codicon (see <https://microsoft.github.io/vscode-codicons/dist/codicon.html>). If using codicons, specify them in the format "$(*icon*)". The icon defaults to a tick if it's not valid. You can also use "better-todo-tree", or "better-todo-tree-filled" if you want to use the icon from the activity view.
+`icon` - used to set a different icon in the tree view. Valid values are octicon names, codicon names in `"$(icon)"` format, `"better-todo-tree"`, and `"better-todo-tree-filled"`. Example: `"bug"` selects the Primer octicon, and `"$(bug)"` selects the VS Code codicon.
 
 `iconColour` - used to set the colour of the icon in the tree. If not specified, it will try to use the foreground colour or the background colour. Colour can be specified as per foreground and background colours, but see note below.
 
@@ -392,7 +392,7 @@ Normally, multiline support is enabled by detecting the use of `\n` in the regex
 **better-todo-tree.regex.regex** (<tt>
 &#x28;&#x2f;&#x2f;&#x7c;&#x23;&#x7c;&#x3c;&#x21;&#x2d;&#x2d;&#x7c;&#x3b;&#x7c;&#x2f;&#x5c;&#x5c;&#x2a;&#x7c;&#x5e;&#x7c;&#x5e;&#x5b;&#x20;&#x5c;&#x5c;&#x74;&#x5d;&#x2a;&#x28;&#x2d;&#x7c;&#x5c;&#x5c;&#x64;&#x2b;&#x2e;&#x29;&#x29;&#x5c;&#x5c;&#x73;&#x2a;&#x28;&#x24;&#x54;&#x41;&#x47;&#x53;&#x29;</tt>)</br>
 
-This defines the regex used to locate TODOs. By default, it searches for tags in comments starting with <tt>&#47;&#47;</tt>, <tt>#</tt>, <tt>;</tt>, <tt>&lt;!--</tt> or <tt>&#47;*</tt>, and also markdown todo lists. This should cover most languages. However if you want to refine it, make sure that the <tt>($TAGS)</tt> is kept as <tt>($TAGS)</tt> will be replaced by the expanded tag list. For some of the extension features to work, <tt>($TAGS)</tt> should be present in the regex, however, the basic functionality should still work if you need to explicitly expand the tag list.
+This defines the regex used to locate TODOs. By default, it searches for tags in comments starting with <tt>&#47;&#47;</tt>, <tt>#</tt>, <tt>;</tt>, <tt>&lt;!--</tt> or <tt>&#47;*</tt>, and markdown todo lists. The <tt>($TAGS)</tt> placeholder is replaced by the expanded tag list. Prefix-only custom regexes that contain <tt>($TAGS)</tt> render text from the detected tag to the end of the physical line.
 
 <sup>*Note: This is a [Rust regular expression](https://docs.rs/regex/1.0.0/regex)</a>, not javascript.*</sup>
 

artifacts/perf/issue28-drop-in-replacement.md

@@ -0,0 +1,74 @@
+# Runtime Benchmarks
+
+- Baseline ref: `a6f60e0ce830c4649ac34fc05e5a1799ec91d151`
+- Current source: working tree
+- Node: `v25.2.0`
+- Selection mode: `scenario-list`
+- Declared suite: `mixed`
+- Result-count validation: `4 rows, suite-consistent=true, all-user-flow=false`
+
+## Machine Profile
+
+| Category | Field | Value |
+| --- | --- | --- |
+| Host | Hostname | n00ne-AERO-17-YD |
+| Host | OS | Ubuntu 22.04.5 LTS |
+| Host | Kernel | 6.8.0-117-generic |
+| Host | Architecture | x64 |
+| Host | Load Average | 3.34, 3.63, 3.14 |
+| Host | Available Parallelism | - |
+| CPU | Model | Intel(R) Core(TM) i9-14900HX |
+| CPU | Vendor | GenuineIntel |
+| CPU | Topology | 16 logical CPU(s), 2 thread(s)/core, 8 core(s)/socket, 1 socket(s), 1 NUMA node(s) |
+| CPU | Frequency | 800 MHz to 5,800 MHz |
+| CPU | Cache | L1d 384 KiB (8 instances), L1i 256 KiB (8 instances), L2 16 MiB (8 instances), L3 36 MiB (1 instance) |
+| Memory | Total RAM | 62.51 GiB (`67,119,767,552 bytes`) |
+| Memory | Available At Collection | 17.92 GiB (`19,236,577,280 bytes`) |
+| Memory | Online Physical RAM | 66.00 GiB (`70,866,960,384 bytes`) |
+| Memory | Swap | total 120 GiB (`128,848,973,824 bytes`); free 109 GiB (`116,994,105,344 bytes`) |
+| Memory | DMI / SPD | Unavailable: /sys/firmware/dmi/tables/smbios_entry_point: Permission denied /dev/mem: Permission denied |
+| Storage | Root Device | nvme0n1 (Samsung SSD 9100 PRO 4TB), 3.64 TiB (`4,000,787,030,016 bytes`), transport nvme, rotational=false, readOnly=false |
+
+## Scenario Model
+
+| Scenario | Kind | User flow | Measurement scope | Input model |
+| --- | --- | --- | --- | --- |
+| scan-large-custom-regex | microbenchmark | - | - | - |
+| attributes-custom-highlight | microbenchmark | - | - | - |
+| open-file-custom-save-rescan-visible-tree | user-flow | Save an already-open file that uses custom regex scanning and redraw the visible tree. | Document save listener, custom-regex document rescan, search-result replacement, and visible-tree render. | Real document text in a VS Code event harness. |
+| tree-render-counts | microbenchmark | - | - | - |
+
+## Metric Model
+
+| Table | Value model | Accuracy model |
+| --- | --- | --- |
+| Latency | Wall-clock elapsed time around each harness flow iteration, summarized as min/p50/p90/p95/max. | Exact for each sampled iteration in this run. |
+| Profiled RSS Burst | Difference between the isolated scenario worker RSS measured immediately before the flow and that worker iteration's OS high-water-mark peak RSS. | Exact for the measured worker iteration, using `process.memoryUsage().rss` at flow start and `process.resourceUsage().maxRSS` for the peak. |
+| Profiled Peak RSS | Highest process RSS reached by each isolated scenario worker iteration. | Exact worker-process high-water mark from `process.resourceUsage().maxRSS`. |
+
+## Latency
+
+| Scenario | Kind | Baseline p50 ms | Current p50 ms | Baseline p90 ms | Current p90 ms | Baseline p95 ms | Current p95 ms |
+| --- | --- | ---: | ---: | ---: | ---: | ---: | ---: |
+| scan-large-custom-regex | microbenchmark | 5.17 | 9.12 | 6.06 | 9.81 | 6.61 | 10.34 |
+| attributes-custom-highlight | microbenchmark | 59.84 | 0.06 | 62.13 | 0.09 | 63.14 | 0.24 |
+| open-file-custom-save-rescan-visible-tree | user-flow | 83.45 | 2.79 | 87.28 | 3.04 | 90.21 | 3.05 |
+| tree-render-counts | microbenchmark | 12.11 | 4.73 | 13.49 | 5.68 | 18.63 | 6.17 |
+
+## Profiled RSS Burst
+
+| Scenario | Kind | Baseline p50 MiB | Current p50 MiB | Baseline p90 MiB | Current p90 MiB | Baseline p95 MiB | Current p95 MiB | Baseline Max MiB | Current Max MiB |
+| --- | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
+| scan-large-custom-regex | microbenchmark | 0.75 | 0.75 | 1.13 | 2.13 | 1.13 | 5 | 1.13 | 5 |
+| attributes-custom-highlight | microbenchmark | 0 | 0 | 0 | 0 | 0 | 0 | 1.63 | 0 |
+| open-file-custom-save-rescan-visible-tree | user-flow | 26.88 | 21 | 27.25 | 21.5 | 27.38 | 21.73 | 27.38 | 21.73 |
+| tree-render-counts | microbenchmark | 0 | 0 | 1.63 | 0 | 4.25 | 0 | 4.25 | 0 |
+
+## Profiled Peak RSS
+
+| Scenario | Kind | Baseline p50 RSS MiB | Current p50 RSS MiB | Baseline p90 RSS MiB | Current p90 RSS MiB | Baseline p95 RSS MiB | Current p95 RSS MiB | Baseline Max RSS MiB | Current Max RSS MiB |
+| --- | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
+| scan-large-custom-regex | microbenchmark | 76.2 | 75.42 | 76.58 | 75.56 | 76.58 | 75.56 | 76.58 | 75.56 |
+| attributes-custom-highlight | microbenchmark | 76.33 | 76.02 | 76.33 | 76.02 | 76.33 | 76.02 | 77.64 | 76.02 |
+| open-file-custom-save-rescan-visible-tree | user-flow | 82.46 | 76.59 | 82.94 | 77.06 | 83.09 | 77.08 | 83.09 | 77.08 |
+| tree-render-counts | microbenchmark | 238.73 | 241.16 | 242.91 | 241.16 | 247.16 | 241.16 | 247.16 | 241.16 |

artifacts/perf/issue28-scan-large-custom-regex.md

@@ -0,0 +1,62 @@
+# Runtime Benchmarks
+
+- Baseline ref: `a6f60e0ce830c4649ac34fc05e5a1799ec91d151`
+- Current source: working tree
+- Node: `v25.2.0`
+- Selection mode: `scenario-list`
+- Declared suite: `microbenchmark`
+- Result-count validation: `1 rows, suite-consistent=true, all-user-flow=false`
+
+## Machine Profile
+
+| Category | Field | Value |
+| --- | --- | --- |
+| Host | Hostname | n00ne-AERO-17-YD |
+| Host | OS | Ubuntu 22.04.5 LTS |
+| Host | Kernel | 6.8.0-117-generic |
+| Host | Architecture | x64 |
+| Host | Load Average | 3.72, 3.72, 3.16 |
+| Host | Available Parallelism | - |
+| CPU | Model | Intel(R) Core(TM) i9-14900HX |
+| CPU | Vendor | GenuineIntel |
+| CPU | Topology | 16 logical CPU(s), 2 thread(s)/core, 8 core(s)/socket, 1 socket(s), 1 NUMA node(s) |
+| CPU | Frequency | 800 MHz to 5,800 MHz |
+| CPU | Cache | L1d 384 KiB (8 instances), L1i 256 KiB (8 instances), L2 16 MiB (8 instances), L3 36 MiB (1 instance) |
+| Memory | Total RAM | 62.51 GiB (`67,119,767,552 bytes`) |
+| Memory | Available At Collection | 17.70 GiB (`19,005,583,360 bytes`) |
+| Memory | Online Physical RAM | 66.00 GiB (`70,866,960,384 bytes`) |
+| Memory | Swap | total 120 GiB (`128,848,973,824 bytes`); free 109 GiB (`116,994,105,344 bytes`) |
+| Memory | DMI / SPD | Unavailable: /sys/firmware/dmi/tables/smbios_entry_point: Permission denied /dev/mem: Permission denied |
+| Storage | Root Device | nvme0n1 (Samsung SSD 9100 PRO 4TB), 3.64 TiB (`4,000,787,030,016 bytes`), transport nvme, rotational=false, readOnly=false |
+
+## Scenario Model
+
+| Scenario | Kind | User flow | Measurement scope | Input model |
+| --- | --- | --- | --- | --- |
+| scan-large-custom-regex | microbenchmark | - | - | - |
+
+## Metric Model
+
+| Table | Value model | Accuracy model |
+| --- | --- | --- |
+| Latency | Wall-clock elapsed time around each harness flow iteration, summarized as min/p50/p90/p95/max. | Exact for each sampled iteration in this run. |
+| Profiled RSS Burst | Difference between the isolated scenario worker RSS measured immediately before the flow and that worker iteration's OS high-water-mark peak RSS. | Exact for the measured worker iteration, using `process.memoryUsage().rss` at flow start and `process.resourceUsage().maxRSS` for the peak. |
+| Profiled Peak RSS | Highest process RSS reached by each isolated scenario worker iteration. | Exact worker-process high-water mark from `process.resourceUsage().maxRSS`. |
+
+## Latency
+
+| Scenario | Kind | Baseline p50 ms | Current p50 ms | Baseline p90 ms | Current p90 ms | Baseline p95 ms | Current p95 ms |
+| --- | --- | ---: | ---: | ---: | ---: | ---: | ---: |
+| scan-large-custom-regex | microbenchmark | 5.39 | 8.74 | 5.99 | 9.96 | 6.96 | 10.51 |
+
+## Profiled RSS Burst
+
+| Scenario | Kind | Baseline p50 MiB | Current p50 MiB | Baseline p90 MiB | Current p90 MiB | Baseline p95 MiB | Current p95 MiB | Baseline Max MiB | Current Max MiB |
+| --- | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
+| scan-large-custom-regex | microbenchmark | 0.75 | 0.75 | 1 | 2.38 | 1.13 | 5.25 | 1.13 | 5.25 |
+
+## Profiled Peak RSS
+
+| Scenario | Kind | Baseline p50 RSS MiB | Current p50 RSS MiB | Baseline p90 RSS MiB | Current p90 RSS MiB | Baseline p95 RSS MiB | Current p95 RSS MiB | Baseline Max RSS MiB | Current Max RSS MiB |
+| --- | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
+| scan-large-custom-regex | microbenchmark | 76.43 | 75.44 | 76.68 | 75.61 | 76.68 | 75.68 | 76.68 | 75.68 |