fix: add retries for flaky integ tests that depend on LLM responses

[Hweinstock]

Issue #, if available: N/A

Description of changes:

Problem

Integration tests in test_session_manager.py that assert on LLM response content are inherently non-deterministic. A model might rephrase or omit expected keywords on any given run,
causing spurious failures that erode CI trust.

Solution

Added pytest-rerunfailures as a dev dependency and gave each potentially flaky test up to 3 total attempts
with a 5-second cooldown. Only the non-deterministic tests are marked; all other tests remain strict single-pass.

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (main@cd2f2a0). Learn more about missing BASE report.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #351   +/-   ##
=======================================
  Coverage        ?   90.92%           
=======================================
  Files           ?       43           
  Lines           ?     4112           
  Branches        ?      634           
=======================================
  Hits            ?     3739           
  Misses          ?      204           
  Partials        ?      169           

Flag	Coverage Δ
unittests	90.92% <ø> (?)

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

☔ 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.