Stabilize large finite time-offset errors

[FlorianPfaff]

Bug

time_offset_error_summary() computed residual magnitudes with np.linalg.norm at the original scale. Large but finite residual components could overflow during squaring, producing inf. _error_stats() then filtered those values out as non-finite and reported zero valid samples.

The same overflow pattern affected mean, standard-deviation, and RMSE reductions, including aggregated time-offset sweeps. For example, two valid residuals equal to 0.75 * np.finfo(float).max were reported as an empty result even though each residual and each expected summary metric is representable.

Fix

• scale residual rows before computing Euclidean norms
• compute means, standard deviations, and RMS values in normalized coordinates before restoring the original scale
• scale aggregation weights and pooled-variance calculations to avoid intermediate overflow
• preserve existing behavior for zero, empty, ordinary-scale, and genuinely non-representable results

Validation

• reproduced the previous inf norm and zero-count summary
• added end-to-end regressions for large finite residual summaries and aggregated sweeps
• focused regression tests: 2 passed
• syntax compilation passed for the implementation and regression test
• compatibility checks passed for offset-grid generation, nearest-time matching, existing aggregate semantics, and recovery of a known offset

Full repository CI is requested by this pull request.

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no	no	23.18s
✅ JSON	prettier	7	0	0	0	0.97s
✅ JSON	v8r	7		0	0	3.38s
✅ MARKDOWN	markdownlint	68	0	0	0	2.0s
✅ MARKDOWN	markdown-table-formatter	68	0	0	0	0.76s
✅ PYTHON	black	1543	197	0	0	90.39s
✅ PYTHON	isort	1543	368	0	0	2.64s
✅ REPOSITORY	betterleaks	yes		no	no	2.3s
✅ REPOSITORY	checkov	yes		no	no	51.86s
✅ REPOSITORY	gitleaks	yes		no	no	13.6s
✅ REPOSITORY	git_diff	yes		no	no	0.46s
✅ REPOSITORY	secretlint	yes		no	no	63.11s
✅ REPOSITORY	syft	yes		no	no	5.56s
✅ REPOSITORY	trivy-sbom	yes		no	no	6.36s
✅ REPOSITORY	trufflehog	yes		no	no	32.3s
✅ YAML	prettier	11	0	0	0	0.76s
✅ YAML	v8r	11		0	0	10.79s
✅ YAML	yamllint	11		0	0	0.61s

Notices

📣 MegaLinter 9.5.0 is out! Discover the new features and security recommendations in the release announcement. (Skip this info by defining SECURITY_SUGGESTIONS: false)

See detailed reports in MegaLinter artifacts

Your project could benefit from a custom flavor, which would allow you to run only the linters you need, and thus improve runtime performances. (Skip this info by defining FLAVOR_SUGGESTIONS: false)

• Documentation: Custom Flavors
• Command: npx mega-linter-runner@9.6.0 --custom-flavor-setup --custom-flavor-linters PYTHON_BLACK,PYTHON_ISORT,COPYPASTE_JSCPD,JSON_V8R,JSON_PRETTIER,MARKDOWN_MARKDOWNLINT,MARKDOWN_MARKDOWN_TABLE_FORMATTER,REPOSITORY_CHECKOV,REPOSITORY_GIT_DIFF,REPOSITORY_GITLEAKS,REPOSITORY_BETTERLEAKS,REPOSITORY_SECRETLINT,REPOSITORY_SYFT,REPOSITORY_TRIVY_SBOM,REPOSITORY_TRUFFLEHOG,YAML_PRETTIER,YAML_YAMLLINT,YAML_V8R

Show us your support by starring ⭐ the repository