test: Deflake Redis RQ metadata timestamp test with a controlled clock

[vdusek]

test_metadata_file_updates in tests/unit/storage_clients/_redis/test_redis_rq_client.py failed on Windows / Python 3.12 with assert accessed_at > accessed_at on identical timestamps (run).

The Redis is_empty() never stamps accessed_at, and get_metadata() reads the record before stamping its own access. The two values the test compared were therefore the stamps of two back-to-back get_metadata() calls (the fixture's purge check and the test's first call), about a millisecond apart, with the asyncio.sleep(0.01) sitting after both. On Windows before Python 3.13, datetime.now() ticks at 1–15.6 ms, so the stamps collide.

The test now drives the mixin's clock through a Mock, uses a read the Redis client does treat as an access (fetch_next_request()), and asserts the exact stamps – no sleeps, no wall clock. Removing the accessed_at update from fetch_next_request() makes it fail, so it still guards the behaviour.

Locally, quantizing the mixin clock to 15.625 ms ticks reproduced the failure 18/20 times (2/40 at 1 ms); the new test passes 20/20 and 40/40 under the same clocks, and the Redis test directory passes 20/20 with -n auto.

Note: is_empty() bumps accessed_at in the memory and file-system clients but not in the Redis one. Left as is here; worth deciding separately whether that's intended.

✍️ Drafted by Claude Code

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 93.73%. Comparing base (8524e23) to head (45ae24a).

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2200      +/-   ##
==========================================
+ Coverage   93.72%   93.73%   +0.01%     
==========================================
  Files         181      181              
  Lines       12825    12825              
==========================================
+ Hits        12020    12022       +2     
+ Misses        805      803       -2     

Flag	Coverage Δ
unit	93.73% <ø> (+0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[Mantisus]

LGTM