Validate Kalman process and measurement covariances

[FlorianPfaff]

Summary

• validate direct Kalman process-noise covariances (Q) as symmetric PSD matrices before prediction
• validate measurement-noise covariances (R) before innovation, standard updates, and robust updates
• reuse the existing backend-native eigenvalue tolerance used for Kalman state covariance validation
• keep failures atomic so invalid noise inputs cannot corrupt the filter state
• retain support for singular positive-semidefinite covariances

Bug

The direct KalmanFilter API previously only received shape/finite checks from the linear-Gaussian primitives. For example, a 1-D filter with P=1 accepted Q=-2 and predict_identity() produced/stored P=-1 because the resulting Gaussian is constructed with check_validity=False. Likewise invalid negative measurement-noise covariance could enter update and innovation calculations.

Tests

Adds regression coverage for:

• non-PSD process noise rejected atomically
• non-PSD measurement noise rejected atomically
• innovation path rejects non-PSD measurement noise
• robust update rejects non-PSD measurement noise atomically
• singular PSD process/measurement noise remains supported

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Max errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no		no	7.33s
✅ JSON	prettier	7	0	0		0	0.62s
✅ JSON	v8r	7		0		0	5.06s
✅ MARKDOWN	markdownlint	68	0	0		0	2.56s
✅ MARKDOWN	markdown-table-formatter	68	0	0		0	0.74s
✅ PYTHON	black	2037	1	0		0	67.35s
✅ PYTHON	isort	2037	6	0		0	2.86s
✅ REPOSITORY	betterleaks	yes		no		no	1.23s
✅ REPOSITORY	checkov	yes		no		no	35.62s
✅ REPOSITORY	git_diff	yes		no		no	0.13s
✅ REPOSITORY	secretlint	yes		no		no	107.41s
✅ REPOSITORY	syft	yes		no		no	2.78s
✅ REPOSITORY	trivy-sbom	yes		no		no	2.74s
✅ YAML	prettier	11	0	0		0	0.58s
✅ YAML	v8r	11		0		0	18.4s
✅ YAML	yamllint	11		0		0	1.01s

Notices

⚠️ Your configuration references items that have been removed from MegaLinter and are ignored: MAKEFILE_CHECKMAKE. See Removed linters to find their replacements.

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@10.0.0 --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_BETTERLEAKS,REPOSITORY_SECRETLINT,REPOSITORY_SYFT,REPOSITORY_TRIVY_SBOM,YAML_PRETTIER,YAML_YAMLLINT,YAML_V8R

Show us your support by starring ⭐ the repository