Parity CI hardening + FitResSummary checkpoint regression guard

[iahncajigas]

Summary

• enforce non-LFS checkout across remaining CI workflows to prevent checkout failures when LFS quota is exhausted
• add a focused regression test for FitResSummaryExamples notebook checkpoints
• add tools/parity/build_line_review_sprint.py and generated parity/line_review_sprint.md to prioritize needs_review topics by lowest line-alignment ratio

Validation

• pytest -q tests/test_fitressummary_notebook_checkpoint.py tests/test_notebook_numeric_checkpoints.py tests/test_notebook_assertions.py
• python tools/parity/build_line_review_sprint.py --top-n 12

Prior evidence runs (green)

• validation-pdf: https://github.com/cajigaslab/nSTAT-python/actions/runs/22607812809
• full-parity-nightly: https://github.com/cajigaslab/nSTAT-python/actions/runs/22607813336

Notes

• this PR keeps the visual validation uniqueness gates introduced earlier and extends checkout hardening to all remaining workflow entry points.

[iahncajigas]

Fresh CI evidence on head :\n- validation-pdf: https://github.com/cajigaslab/nSTAT-python/actions/runs/22620617812 (success)\n- full-parity-nightly: https://github.com/cajigaslab/nSTAT-python/actions/runs/22620618645 (success)

[iahncajigas]

Fresh CI evidence on branch head b0e3a1d:

• validation-pdf: https://github.com/cajigaslab/nSTAT-python/actions/runs/22620617812 (success)
• full-parity-nightly: https://github.com/cajigaslab/nSTAT-python/actions/runs/22620618645 (success)

[iahncajigas]

Closing: superseded by v0.3.0 merge (PR #50) and subsequent work. The codex-generated fixtures, CI workflows, and demo examples were replaced by hand-curated implementations. All 484 Matlab methods are now verified in PORTING_MAP.md.