Stabilize measurement reliability covariance scaling

[FlorianPfaff]

Summary

• evaluate reliability covariance-scale overflow in log space before exponentiation
• apply a finite max_scale cap before the power operation can underflow to zero
• raise a clear ValueError when an uncapped scale cannot be represented as a finite float
• add regressions for both capped and uncapped extreme exponents

Bug

reliability_to_covariance_scale() computed

1.0 / max(reliability, floor) ** exponent

before applying max_scale. For large but finite exponents, the denominator can underflow to zero. The public helper then either returned inf or raised ZeroDivisionError; an explicit finite max_scale could not protect the computation because it was applied too late.

Fix

Compute the logarithm of the requested scale first. If it reaches a supplied finite cap, return that cap without evaluating the unstable power. If an uncapped result exceeds the largest finite float, raise a descriptive ValueError rather than leaking an incidental arithmetic exception or infinity.

Normal representable inputs still use the existing power expression, preserving current results and rounding behavior.

Validation

• reproduced the existing ZeroDivisionError with reliability=0.5 and exponent=1e6
• verified the patched calculation returns 7.0 when max_scale=7.0
• verified the uncapped case raises ValueError containing covariance scale overflows
• branch is two commits ahead of and zero commits behind current main
• diff is limited to one production function and one focused regression file

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no	no	21.33s
✅ JSON	prettier	7	0	0	0	1.55s
✅ JSON	v8r	7		0	0	3.41s
✅ MARKDOWN	markdownlint	68	0	0	0	1.89s
✅ MARKDOWN	markdown-table-formatter	68	0	0	0	0.78s
✅ PYTHON	black	1490	183	0	0	85.42s
✅ PYTHON	isort	1490	328	0	0	2.47s
✅ REPOSITORY	betterleaks	yes		no	no	2.34s
✅ REPOSITORY	checkov	yes		no	no	50.14s
✅ REPOSITORY	gitleaks	yes		no	no	13.59s
✅ REPOSITORY	git_diff	yes		no	no	0.38s
✅ REPOSITORY	secretlint	yes		no	no	58.89s
✅ REPOSITORY	syft	yes		no	no	6.51s
✅ REPOSITORY	trivy-sbom	yes		no	no	5.61s
✅ REPOSITORY	trufflehog	yes		no	no	30.87s
✅ YAML	prettier	11	0	0	0	0.67s
✅ YAML	v8r	11		0	0	10.37s
✅ YAML	yamllint	11		0	0	0.64s

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