[PM-36604] test: Fix flaky WatchServiceTests

[KatherineInCode]

🎟️ Tracking

PM-36604

📔 Objective

Fixes three flaky tests in WatchServiceTests:

1. syncWithWatch_shouldNotConnect_sendsNeedSetup — captures the application context inside block 2's closure rather than reading the shared mock property afterwards. A still-running syncCiphersTask from the first sync could overwrite updateApplicationContextReceivedApplicationContext between the resume() call and the assertion.
2. syncWithWatch_vaultLocked_skipsCipherSync — replaces Task.sleep(10_000_000) with a deterministic barrier sync: after the locked event, a shouldConnect=false event triggers a .needSetup response (which short-circuits before decryptCiphers) and provides a reliable signal that the locked sync was fully processed with no extra decryption.
3. syncWithWatch_vaultUnlocked_syncsAfterUnlock — replaces Task.sleep(10_000_000) + post-sleep assertion with an in-closure assertion on updateApplicationContextCallsCount. The locked sync returns early at the isVaultLocked guard without calling updateApplicationContext, so asserting callsCount == 1 inside the unlock callback is both correct and deterministic.

[github-actions]

🤖 Bitwarden Claude Code Review

Overall Assessment: APPROVE

Reviewed test-only changes to WatchServiceTests.swift that fix three flaky tests by removing fixed-duration Task.sleep calls and a TOCTOU race on a shared mock property. The fixes use deterministic patterns already established in the file: capturing values inside the callback closure, and using a subsequent event as a barrier sync that guarantees the prior event's task has been processed. Traced the production DefaultWatchService.syncWithWatch(userId:shouldConnect:isVaultLocked:) flow and confirmed the locked-branch early return (before updateApplicationContext) makes the test assumptions correct. The previously flagged experimental time-limit tests were removed in commit 0a5b6ae9 and are not present in the final diff.

Code Review Details

No findings.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 87.31%. Comparing base (d61c576) to head (e16f245).
⚠️ Report is 20 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2617   +/-   ##
=======================================
  Coverage   87.31%   87.31%           
=======================================
  Files        1912     1912           
  Lines      170379   170400   +21     
=======================================
+ Hits       148759   148779   +20     
- Misses      21620    21621    +1     

☔ View full report in Codecov by Sentry.
📢 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.