Decouple the graph language from the runtime

[akashgit]

What

Breaks the two runtime couplings in the graph language so factory.workflow imports with zero runtime deps (pydantic + stdlib only):

1. primitives.py inlines the tiny ProjectState enum and replaces the FactoryConfig field type on Factory with dict[str, Any] (it was never read by the language).
2. definitions.py imports ProjectState from primitives instead of factory.models.
3. __init__.py no longer re-exports WorkflowExecutor / ExecutionResult — DSH is the runtime, not the executor.

Why

This is the first step of extracting refactory's graph semantics into a standalone language package (the PyTorch-analogy split: refactory = semantics, DSH = runtime). The runtime files (executor/skill_export/llm_loop/… ) still live in the repo and still work, but they are no longer reachable through the language import.

Tests

• pytest tests/test_workflow_primitives.py tests/test_workflow_definitions.py — 95 passed
• pytest tests/test_outer_loop/ — 563 passed
• SRF graph suite — 496 passed, 1 skipped

[github-actions]

Sentrux Quality Report

Absolute

Scanning ....
[scan] git ls-files: 651 total, 637 kept, 14 dropped (ext:14, meta:0, big:0)
[build_project_map] 637 files, 108 unique dirs, 100 cache misses, 5.3ms
[resolve] 1293 resolved, 1680 unresolved (of 2973 total specs)
[resolve_imports] project_map 5.4ms, suffix_idx 1.2ms, suffix_resolve 20.0ms, total 26.6ms
[build_graphs] 637 files | maps 2.4ms, imports 26.8ms, calls+inherit 7.0ms, total 36.2ms | 1292 import, 10486 call, 4 inherit edges
sentrux check — 3 rules checked

Quality: 4400

✗ [Error] max_cc: 9 function(s) exceed max cyclomatic complexity of 30
    factory/workflow/executor.py:_execute_data (cc=53)
    factory/outer_loop/reflector.py:_extract_eval_patterns (cc=46)
    factory/cli/_ceo_helpers.py:_validate_ceo_flags (cc=43)
    factory/cli/_ceo_helpers.py:_execute_ceo (cc=43)
    factory/outer_loop/mutations.py:_try_mutation (cc=40)
    examples/chess_evolve.py:main (cc=33)
    factory/cli/_task_builder.py:_build_ceo_task (cc=32)
    factory/outer_loop/reflector.py:_llm_reflect (cc=32)
    factory/outer_loop/mutations.py:validate_and_repair (cc=31)

✗ 1 violation(s) found

Diff (vs base branch)

Scanning ....
[scan] git ls-files: 651 total, 637 kept, 14 dropped (ext:14, meta:0, big:0)
[build_project_map] 637 files, 108 unique dirs, 100 cache misses, 5.4ms
[resolve] 1293 resolved, 1680 unresolved (of 2973 total specs)
[resolve_imports] project_map 5.5ms, suffix_idx 1.2ms, suffix_resolve 23.0ms, total 29.7ms
[build_graphs] 637 files | maps 2.6ms, imports 29.9ms, calls+inherit 7.6ms, total 40.2ms | 1292 import, 10486 call, 4 inherit edges
sentrux gate — structural regression check

Quality:      4415 -> 4400
Coupling:     0.81 → 0.81
Cycles:       3 → 4
God files:    3 → 3

Distance from Main Sequence: 0.35

✗ DEGRADED
  ✗ Cycles increased: 3 → 4

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 83.76%. Comparing base (f0d28a0) to head (7ad78fe).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1500   +/-   ##
=======================================
  Coverage   83.75%   83.76%           
=======================================
  Files         225      225           
  Lines       25757    25760    +3     
  Branches     4209     4209           
=======================================
+ Hits        21574    21578    +4     
  Misses       3201     3201           
+ Partials      982      981    -1     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.