fix: avoid SQLite variable limit in thread lookups

[steipete]

Large neighbors --limit requests fail while hydrating results because ThreadsByIDs binds one SQL parameter per ID plus the repository. More than 32,765 result IDs exceed the bundled SQLite limit.

Pass the IDs through one JSON array parameter using SQLite's existing json_each support. This preserves repository scoping, duplicate and missing-ID behavior, and result ordering in callers, and also fixes the shared lookup used by search and cluster views. No schema or CLI changes are needed.

The regression reproduces the previous too many SQL variables error with 32,768 IDs and now passes, including duplicate IDs, missing IDs, and an ID from another repository. Updated the neighbor documentation and Unreleased changelog. Thanks @masatohoshino for reporting this in #238.

This fixes the SQL-limit portion of #238. The proposed temporal filters and metadata in #239 remain open under their existing product-decision hold.

Validation: focused store regression tests; full make check on AWS Crabbox; independent Codex review through P3 with no findings. PR CI covers Linux, macOS, Windows, and documentation.

[clawsweeper]

🦞👀
ClawSweeper picked this up.

Pull request received. I will update this pull request when review starts.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-16T21:23:54.908434Z	2a9fee9	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.