feat(mcp): add file_history tool with per-file patch events

[samzong]

What's changed?

• Add a read-only file_history MCP tool: reports which indexed sessions touched a file, with session id, source, project, title, kind, name, target, event_seq, truncated summary, and event timestamp.
• Match targets bidirectionally at / or \\ boundaries so relative and absolute forms of the same path find each other.
• Normalize patch writes into per-file file_write events: Codex apply_patch calls previously produced target-less tool_call events, and OpenCode multi-file patches kept only the first file. Event parser versions bumped (codex 1→2, opencode 2→3) so existing sessions backfill on the next sync.
• Order events by COALESCE(e.timestamp, s.updated_at, s.started_at) so sources that omit event timestamps (Cursor) still surface by session activity.

Why

• Agents need a fast answer to "which sessions touched this file" that full-text search cannot express; without the patch normalization the tool would systematically miss Codex file writes, its primary write path.

Verification

• make check (audit → fmt → clippy -D warnings → workspace tests incl. bench feature) passes.
• New tests cover bidirectional suffix matching, per-file patch events for Codex and OpenCode, timestamp-less event ordering, limit clamping, and that attrs_json never leaks into tool output.

[codspeed-hq]

Merging this PR will improve performance by 13.35%

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 1 improved benchmark
✅ 21 untouched benchmarks
⏩ 2 skipped benchmarks¹

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
⚡	export_jsonl	21.4 ms	18.9 ms	+13.35%

Tip

Curious why performance improved? Comment @codspeedbot explain why performance improved on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing feat/mcp-file-history (8b2e343) with main (c4f8af3)²}

Footnotes

1. 2 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

2. No successful run was found on main (a45c10e) during the generation of this report, so c4f8af3 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩