Avoid overflow when validating SO(3) product particles

[FlorianPfaff]

Summary

• replace the raw Euclidean-norm nonzero check for SO(3) product particles with a componentwise nonzero check
• preserve rejection of zero quaternions while allowing the existing scale-first quaternion normalizer to handle extreme finite magnitudes
• add a regression using backend-dtype quaternions with components at half the largest finite value

Bug

SO3ProductParticleFilter._as_particle_array checked whether a quaternion was nonzero using linalg.norm. For a valid finite quaternion such as [float_max / 2, float_max / 2, 0, 0], the intermediate sum of squares overflows. Under strict NumPy floating-point handling this raises before PyRecEst's existing overflow-safe normalize_quaternions helper can run.

The quaternion's scale is irrelevant to the represented rotation, so rejecting it is incorrect.

Fix

Use a componentwise nonzero predicate for validation. This avoids squaring the input and leaves normalization to the existing scale-first helper.

Validation

• isolated reproduction: the old raw norm raises FloatingPointError: overflow encountered in multiply
• patched path: the same finite quaternion normalizes to [1/sqrt(2), 1/sqrt(2), 0, 0] with unit norm under strict floating-point settings
• focused Ruff, documentation, security, dependency-review, and package smoke checks passed on the reviewed diff before history cleanup
• the branch now contains one intentional commit; repository CI is rerunning on that final head

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no	no	18.94s
✅ JSON	prettier	7	0	0	0	0.72s
✅ JSON	v8r	7		0	0	3.02s
✅ MARKDOWN	markdownlint	68	0	0	0	1.7s
✅ MARKDOWN	markdown-table-formatter	68	0	0	0	0.93s
✅ PYTHON	black	1866	88	0	0	69.3s
✅ PYTHON	isort	1866	140	0	0	2.13s
✅ REPOSITORY	betterleaks	yes		no	no	1.51s
✅ REPOSITORY	checkov	yes		no	no	42.19s
✅ REPOSITORY	gitleaks	yes		no	no	13.5s
✅ REPOSITORY	git_diff	yes		no	no	0.26s
✅ REPOSITORY	secretlint	yes		no	no	60.68s
✅ REPOSITORY	syft	yes		no	no	4.18s
✅ REPOSITORY	trivy-sbom	yes		no	no	4.23s
✅ REPOSITORY	trufflehog	yes		no	no	24.26s
✅ YAML	prettier	11	0	0	0	0.56s
✅ YAML	v8r	11		0	0	9.32s
✅ 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