fix(search): apply cosine radius across iterative recall paths

[knqiufan]

Summary

Fixes #394.

agentic and llm_multiround accepted radius but bypassed the manager's dense-recall filter. A request with dense hits at cosine 0.9 and 0.2 and radius=0.5 now removes the 0.2 hit before fusion on every iterative query. Existing vector and hybrid filtering remains in place.

• Share request-default resolution and dense filtering across search orchestration. Resolve the default before capping top_k: explicit values win (including zero), unlimited searches default to 0.5, and positive top_k has no implicit floor.
• Pass the resolved floor through user/agent AGENTIC and LLM_MULTIROUND. Filter both atomic-fact and episode-subject dense hits before user AGENTIC MaxSim pooling, and filter each multi-round dense pool before RRF.
• Remove the three inert RankInput.radius arguments. Never map cosine thresholds to ahybrid_retrieve(min_score=...), which would compare them with a different score scale.
• Document the dense-only contract and the episode HYBRID-only scope of min_score. Add 49 regression cases covering dispatch/defaults, boundaries, sparse preservation, empty dense pools, both agent kinds, and repeated queries with real RRF/MaxSim operators.

Area

• Architecture method
• Benchmark
• Use case
• Documentation
• Developer experience
• CI, build, or release

Verification

Pre-PR Linux validation uses the same source/test tree as this PR; validation branches change only workflow configuration:

• Linux CI: make lint and make package passed; make test passed on Python 3.12 and 3.13 (2599 passed, 4 skipped each); make integration passed on both versions (183 passed, 5 skipped, 7 deselected each).
• Local search suite: 310 passed. Windows required an import-only fcntl stand-in that raises on any attempted POSIX operation; the Linux runs above use the real platform without that stand-in.
• Documentation links, Conventional Commit format, and git diff --check passed. OpenAPI export has no drift.
• Milvus 2.6.22 validation: repository contract (5 passed), /get telemetry (1 passed), tiered API end-to-end tests (27 passed), and both-backend index contract (18 passed). The first run could not pull the upstream compose file's minio/minio image; the successful validation uses the same MinIO version tag from quay.io/minio/minio. This environment-only workaround is not part of the PR; the unchanged upstream workflow may encounter the same Docker Hub pull failure.

Checklist

• I kept the change scoped to the relevant area.
• I am opening this from a separate branch, not pushing directly to main.
• I updated docs, examples, or setup notes when behavior changed.
• I added or updated tests when the change affects behavior.
• I did not commit secrets, .env files, dependency folders, or generated output.
• Active relative links in Markdown files resolve.

Notes for Reviewers

This implements option (a) from the issue: a pre-fusion dense cosine floor. Keyword matches, cluster expansions, and case-to-skill linkage remain independent recall routes; radius is not a final-score floor or a promise of monotonically decreasing result counts. The existing min_score behavior is unchanged.

The intended behavior change includes default unlimited iterative searches: they now honor the documented 0.5 floor. Internal helper callers that omit the new optional argument retain unfiltered behavior. VECTOR and HYBRID already filtered dense hits before this change; the unused ranker argument was not evidence that those paths ignored radius.

By submitting this pull request, I agree that my contribution is licensed under the Apache License 2.0.