Document repeat reliability methodology

[bernalde]

Summary

• Document how repeat reliability changes Window Sticker methodology claims.
• Separate cross-instance bootstrap uncertainty from per-instance repeat-count reliability.
• Clarify that Noori et al. 2026 analytic guarantees apply to Bernoulli success events, R_c, RTT/TTS, CETS, and thresholded continuous metrics, while continuous Response and PerfRatio curves remain bootstrap-based unless thresholded.
• Add a methodology criticism mapping and documentation regression tests.

Tests run

• python -m pytest tests/test_methodology_docs.py -q - 2 passed.
• python run_tests.py unit --fast - 413 passed, 12 warnings.
• flake8 src --count --select=E9,F63,F7,F82 --show-source --statistics - 0 critical issues.
• flake8 src --count --exit-zero --max-complexity=10 --max-line-length=120 --statistics - exited 0; printed existing advisory style warnings.
• python run_tests.py integration - 11 passed.

Notes about tests not run

• Did not run local tutorial notebook execution or local coverage report generation; CI covers those heavier checks.
• No separate type-check or formatting command is documented for this repository.

Closes #76

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[bernalde]

No blocking issues found.

1. Blocking issues: none.
2. Nonblocking issues: none.
3. Questions: none.
4. Tests run and outcomes:
   • python -m pytest tests/test_methodology_docs.py -q - 2 passed.
   • python run_tests.py all --fast - unit, integration, and smoke suites passed.
   • flake8 src --count --select=E9,F63,F7,F82 --show-source --statistics - 0 critical issues.
   • flake8 src --count --exit-zero --max-complexity=10 --max-line-length=120 --statistics - exited 0 with existing advisory style warnings.
   • gh pr checks 81 --repo usra-riacs/stochastic-benchmark - all reported checks passed.
5. Merge recommendation: merge-ready. I am submitting this as COMMENT because GitHub does not allow the authenticated author account to approve its own PR.