chore: retry integration tests and generate report

[njbrake]

Description

Integration tests fail because of the provider overload/model flakiness, etc. Lets add a little retry logic on our side to smooth that out. This should also generate a report so you can see which ones needed the retry

PR Type

• 🚦 Infrastructure

Relevant issues

Checklist

• I understand the code I am submitting.
• I have added unit tests that prove my fix/feature works
• I have run this code locally and verified it fixes the issue.
• New and existing tests pass locally
• Documentation was updated where necessary
• I have read and followed the contribution guidelines
• AI Usage:
  • No AI was used.
  • AI was used for drafting/refactoring.
  • This is fully AI-generated.

AI Usage Information

• AI Model used: Opus 4.5
• AI Developer Tool used: CC + AoE
• Any other info you'd like to share:

Claude did an analysis of options:

  Recommended Solution: pytest-rerunfailures + Custom GitHub Actions Reporter

  I recommend using pytest-rerunfailures (the most mature solution with 2.2M weekly downloads) combined with a custom conftest.py plugin for GitHub Actions summary reporting.

  Why pytest-rerunfailures?
  ┌──────────────────────┬──────────────────────┬──────────────┬──────────────┐
  │       Criteria       │ pytest-rerunfailures │ pytest-retry │    flaky     │
  ├──────────────────────┼──────────────────────┼──────────────┼──────────────┤
  │ Weekly downloads     │ 2.2M                 │ ~Low         │ Moderate     │
  ├──────────────────────┼──────────────────────┼──────────────┼──────────────┤
  │ Health score         │ 85/100               │ 50/100       │ N/A          │
  ├──────────────────────┼──────────────────────┼──────────────┼──────────────┤
  │ pytest-xdist support │ Yes (crash recovery) │ No           │ Partial      │
  ├──────────────────────┼──────────────────────┼──────────────┼──────────────┤
  │ Active maintenance   │ Yes (Oct 2025)       │ Limited      │ Moderate     │
  ├──────────────────────┼──────────────────────┼──────────────┼──────────────┤
  │ Exception filtering  │ Yes                  │ Yes          │ Via callback │
  ├──────────────────────┼──────────────────────┼──────────────┼──────────────┤
  │ Per-test config      │ Yes                  │ Yes          │ Yes          │
  └──────────────────────┴──────────────────────┴──────────────┴──────────────┘
 . ```

When answering questions by the reviewer, please respond yourself, do not copy/paste the reviewer comments into an AI system and paste back its answer. We want to discuss with you, not your AI :) 

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
see 45 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[njbrake]

Integration tests:

https://github.com/mozilla-ai/any-llm/actions/runs/21448115184/job/61769393350

👨‍🍳

[daavoo]

❤️