Isolate integration tests per run to fix concurrent-run flakiness

[AndreyVMarkelov]

Problem

The integration tests all shared a single hard-coded folder (/Testing/Dropbox.Api.Tests) on one shared Dropbox account. Each test's Initialize asserted the folder was empty and Cleanup emptied it, so overlapping CI runs (e.g. several open PRs building at once) raced on the folder's contents.

This surfaced as intermittent, non-reproducible failures such as:

• Assert.AreEqual failed. Expected:<0>. Actual:<1>. in Initialize
• entry-count mismatches (Expected:<3>. Actual:<2>)
• path_lookup/not_found during Cleanup

max-parallel: 1 only serializes the OS matrix within a single workflow run — it does nothing across separate workflow runs, so concurrent PR builds collided.

Fix

Give each test run its own unique subfolder /Testing/Dropbox.Api.Tests/<guid>, created in ClassInitialize and removed in a new ClassCleanup. Concurrent runs can no longer see or delete each other's files.

Validation (local)

• Sequential runs (5×) and single runs: 20/20 pass
• Two concurrent full suites: previously both failed deterministically on the folder race; after the fix that race is gone (a run that would have failed now passes).
• dotnet build clean; dotnet format --verify-no-changes --severity warn passes.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 53.81%. Comparing base (f2a7feb) to head (4bd1ac9).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #378   +/-   ##
=======================================
  Coverage   53.81%   53.81%           
=======================================
  Files          16       16           
  Lines        1403     1403           
  Branches      134      134           
=======================================
  Hits          755      755           
  Misses        608      608           
  Partials       40       40           

Flag	Coverage Δ
integration	50.32% <ø> (ø)
unit	10.54% <ø> (ø)

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.