Reject masked assignment inputs

[FlorianPfaff]

What changed

• detect genuinely masked NumPy values before backend conversion
• reject masked assignment counts, cost matrices, and row/column non-assignment costs with explicit validation errors
• preserve support for fully unmasked MaskedArray inputs
• add regression coverage for both assignment solvers and nested object-array inputs

Root cause

Backend/NumPy array conversion can discard MaskedArray mask metadata and expose the hidden payload. In the assignment utilities, a missing cost could therefore be interpreted as a real numeric cost and silently change the selected assignment; a masked k could likewise be converted into an unintended count.

Validation

• reproduced the pre-fix behavior: masked matrix entries were accepted and selected using their hidden values
• focused regression suite: 4 tests passed
• source and test files compile successfully
• GitHub CI should run the repository's full supported backend/test matrix on this PR

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no	no	25.44s
✅ JSON	prettier	7	0	0	0	1.17s
✅ JSON	v8r	7		0	0	4.68s
✅ MARKDOWN	markdownlint	68	0	0	0	1.76s
✅ MARKDOWN	markdown-table-formatter	68	0	0	0	1.03s
✅ PYTHON	black	1810	57	0	0	89.41s
✅ PYTHON	isort	1810	89	0	0	2.69s
✅ REPOSITORY	betterleaks	yes		no	no	2.17s
✅ REPOSITORY	checkov	yes		no	no	54.74s
✅ REPOSITORY	gitleaks	yes		no	no	16.27s
✅ REPOSITORY	git_diff	yes		no	no	0.35s
✅ REPOSITORY	secretlint	yes		no	no	80.94s
✅ REPOSITORY	syft	yes		no	no	5.76s
✅ REPOSITORY	trivy-sbom	yes		no	no	6.32s
✅ REPOSITORY	trufflehog	yes		no	no	32.41s
✅ YAML	prettier	11	0	0	0	0.76s
✅ YAML	v8r	11		0	0	11.45s
✅ YAML	yamllint	11		0	0	0.54s

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

[FlorianPfaff]

Closing as low value. Recursively propagating NumPy masked-array semantics through the assignment solvers adds substantial complexity for an input type that is not otherwise supported consistently by the solver stack. A broader API-level design would be needed before adopting this behavior.