[resolve #11] /review finding by doc-scribe (report: .temp/output-review-from-inference-rle-mask-2026-04-14.md):
"Changelog entry for rle_to_mask / mask_to_rle relocation doesn't mention the dtype change from uint8 to bool..."

---
Co-authored-by: Claude Code <noreply@anthropic.com>

AGENTS.md requires at least a one-line docstring per test function.

[resolve roboflow#11] /review finding by qa-specialist (report: .temp/output-review-fix-coco-loader-path-containment-2026-04-17.md):
"Missing docstrings on new test functions — AGENTS.md convention violated"

---
Co-authored-by: Claude Code <noreply@anthropic.com>

- Rename TestResize → TestCompactMaskResize for naming consistency with
  other test classes in test_compact_mask.py
- Add test_join_true_true_junction_no_zero_run targeting the True/True
  boundary merge introduced in the prior fix
- Move _PARALLEL_THRESHOLD to module level alongside _L3_DENSITY_THRESHOLD
- Cap ThreadPoolExecutor max_workers=min(n, os.cpu_count() or 4) to avoid
  thread contention in library code
- Document identity-resize shallow-copy semantics in inline comment

[resolve #3] @Copilot (gh): "TestResize is a very generic class name..."
[resolve #9] /review by sw-engineer: "_PARALLEL_THRESHOLD constant buried in method body"
[resolve #10] /review by sw-engineer: "ThreadPoolExecutor() with no max_workers argument"
[resolve #11] /review by sw-engineer: "identity resize does shallow copy of list(self._rles)"
[resolve #12] /review by qa-specialist: "_rle_join_cols has no test for True/True junction"

---
Co-authored-by: Claude Code <noreply@anthropic.com>

- [resolve roboflow#3] with_nmm: for OBB merge groups with >1 member, set merged
  xyxy from winner's OBB corners instead of union AABB of all members.
  Challenge: evidence=VALID suggestion=VALID resolution=as-suggested
- [resolve roboflow#6] with_nms + with_nmm docstrings: document three-path dispatch
  order (mask → OBB → AABB) so callers know OBB is now routed differently.
  Challenge: evidence=VALID suggestion=VALID resolution=as-suggested
- [resolve roboflow#11] with_nmm: cast OBB to float32 (was float64), aligning with
  with_nms; relax _group_overlapping_oriented_boxes annotations to np.floating.
  Challenge: evidence=VALID suggestion=VALID resolution=as-suggested

[resolve group] PR roboflow#2303 — items 3, 6, 11

---
Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>
Co-authored-by: OpenAI Codex <codex@openai.com>