fix(honcho): include file size in memo key to detect sub-second honch…

[giggling-ginger]

What does this PR do?

_extract_honcho_cache_busting_config memoized its result using (path, mtime_ns) as the cache key. On most filesystems, mtime_ns has 1-second granularity, so rapid consecutive writes to honcho.json within the same second would share an identical mtime_ns. The second call would then hit the stale cache and return the old configuration, causing pinPeerName flips to be ignored.

This PR adds st_size to the memoization key. When pinPeerName flips (true ↔ false), the serialized JSON changes length by one byte, ensuring the cache key changes even when mtime_ns remains identical. Since st_size is retrieved during the existing path.stat() call, there is no additional I/O cost.

Related Issue

Fixes #47844

Type of Change

• 🐛 Bug fix (non-breaking change that fixes an issue)
• ✨ New feature (non-breaking change that adds functionality)
• 🔒 Security fix
• 📝 Documentation update
• ✅ Tests (adding or improving test coverage)
• ♻️ Refactor (no behavior change)
• 🎯 New skill (bundled or hub)

Changes Made

• Modified gateway/run.py in _extract_honcho_cache_busting_config: Included st_size in the memo_key tuple to ensure cache invalidation on file content changes that occur within the same second.

How to Test

1. Run the specific test case that reproduces the race condition:
   scripts/run_tests.sh tests/honcho_plugin/test_pin_peer_name.py::TestPinTransition::test_cache_busting_signature_reflects_pin_peer_name
2. Verify that the test passes (it previously failed due to identical cache signatures).
3. Confirm all tests in the file pass: scripts/run_tests.sh tests/honcho_plugin/test_pin_peer_name.py

Checklist

Code

• I've read the Contributing Guide
• My commit messages follow Conventional Commits
• I searched for existing PRs to make sure this isn't a duplicate
• My PR contains only changes related to this fix/feature
• I've run pytest tests/ -q and all tests pass
• I've added tests for my changes (this PR addresses an existing test failure)
• I've tested on my platform: Ubuntu 22.04

Documentation & Housekeeping

• I've updated relevant documentation — N/A
• I've updated cli-config.yaml.example — N/A
• I've updated CONTRIBUTING.md or AGENTS.md — N/A
• I've considered cross-platform impact — Yes, path.stat() is standard.
• I've updated tool descriptions/schemas — N/A

Screenshots / Logs

[100.0% | 43/43 | ✓43 | ✗ 0] ✓ tests/honcho_plugin/test_pin_peer_name.py (43✓, 2.2s)

[alt-glitch]

Duplicate of #46385 — both PRs add st_size to the _extract_honcho_cache_busting_config memo key in gateway/run.py to bust the cache on coarse-mtime filesystems when honcho.json is rewritten within the same second. #46385 (open, earlier) is the canonical fix; the mechanism here is identical.

[giggling-ginger]

Duplicate of #46385 — both PRs add st_size to the _extract_honcho_cache_busting_config memo key in gateway/run.py to bust the cache on coarse-mtime filesystems when honcho.json is rewritten within the same second. #46385 (open, earlier) is the canonical fix; the mechanism here is identical.

Thanks for the context, alt-glitch. I see that #46385 covers the same logic. Glad to see the issue is being addressed with the same approach. I'll close this PR.