Remove bloom prefilter application in cudf-polars streaming

[wence-]

Description

If join filter pushdown is applied, additionally applying a bloom filter does additional work for no benefit. Bloom filters are also only on immediate local inputs to a join.

In an upcoming change we will change the join filter pushdown to be a hint that can at runtime decide how to apply itself. To prepare for that, remove the special case code that just applies bloom filters: it will be subsumed by the upcoming changes.

Checklist

• I am familiar with the Contributing Guidelines.
• New or existing tests cover these changes.
• The documentation is up to date with these changes.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 99e85c69-aadb-48af-93ac-4a7ad633d3d9

📥 Commits

Reviewing files that changed from the base of the PR and between 777fcd5 and 4dff822.

📒 Files selected for processing (7)

• python/cudf_polars/cudf_polars/streaming/actor_graph/collectives/common.py
• python/cudf_polars/cudf_polars/streaming/actor_graph/join.py
• python/cudf_polars/cudf_polars/streaming/actor_graph/utils.py
• python/cudf_polars/cudf_polars/utils/config.py
• python/cudf_polars/tests/streaming/test_join.py
• python/cudf_polars/tests/streaming/test_tracing.py
• python/cudf_polars/tests/test_config.py

💤 Files with no reviewable changes (4)

• python/cudf_polars/cudf_polars/streaming/actor_graph/utils.py
• python/cudf_polars/tests/streaming/test_tracing.py
• python/cudf_polars/cudf_polars/utils/config.py
• python/cudf_polars/tests/test_config.py

---

📝 Walkthrough

Summary by CodeRabbit

• Removed Features

  • Removed the dynamic join bloom-filter prefiltering step.
  • Removed configuration options for join prefilter thresholds, key limits, and tracing.
  • Dynamic joins now use three collective operations for allgather and left/right shuffles.

• Behavior Changes

  • Shuffle joins now proceed directly through both shuffles and chunkwise joining.
  • Simplified dynamic-planning validation and configuration behavior.

• Tests

  • Removed obsolete prefilter and tracing coverage.
  • Added coverage for join filter pushdown configuration and validation.

Walkthrough

Dynamic-planning joins no longer allocate or execute Bloom-filter prefiltering. Related configuration, tracing metadata propagation, and tests were removed or updated.

Changes

Dynamic join prefilter removal

Layer / File(s)	Summary
Simplify dynamic shuffle joins python/cudf_polars/cudf_polars/streaming/actor_graph/collectives/common.py, python/cudf_polars/cudf_polars/streaming/actor_graph/join.py, python/cudf_polars/tests/streaming/test_join.py	Shuffle joins now use two shuffle channels and three collective IDs. Prefilter decisions, filter tasks, and related parameters were removed.
Remove prefilter configuration python/cudf_polars/cudf_polars/utils/config.py, python/cudf_polars/tests/test_config.py	Dynamic-planning options no longer expose or validate join-prefilter settings. Configuration tests now cover join-filter-pushdown settings.
Update tracing and test coverage python/cudf_polars/cudf_polars/streaming/actor_graph/utils.py	Actor shutdown logging no longer copies tracer.extra metadata into the record.

Estimated code review effort: 3 (Moderate) | ~25 minutes

Possibly related PRs

• rapidsai/cudf#22996: Introduces related dynamic join-prefilter functionality that this PR removes.

Suggested labels: improvement, non-breaking

Suggested reviewers: tomaugspurger, vyasr, matt711

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and concisely describes the removal of bloom prefilter application from cudf-polars streaming.
Description check	✅ Passed	The description explains why bloom prefilter application is removed and how the change supports future join filter pushdown behavior.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches 💡 1

🛠️ Fix failing CI checks 💡

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

_{Comment @coderabbitai help to get the list of available commands.}

[TomAugspurger]

I think this is fine, but just flagging that this configuration change is breaking (at least currently, perhaps the subsequent PR will restore this though I'd guess it'll use different names).

[wence-]

/merge