Validate Autograd multinomial pvals

[FlorianPfaff]

Summary

• stop re-exporting the Autograd backend's raw multinomial sampler directly
• validate pvals before dispatch so non-real probability vectors fail with a clear TypeError
• add focused Autograd backend regression coverage

Bug fixed

The Autograd backend still exposed autograd.numpy.random.multinomial without the probability validation added for the NumPy backend. Inputs with non-real probability dtypes could therefore be passed through to the low-level sampler instead of being rejected consistently by PyRecEst's backend contract.

Validation

• Syntax-checked the modified source and new test file with ast.parse locally.
• Full pytest was not run locally because the sandbox cannot resolve github.com to clone/install the repository dependency set.

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no	no	73.52s
✅ JSON	prettier	7	0	0	0	0.87s
✅ JSON	v8r	7		0	0	5.15s
✅ MARKDOWN	markdownlint	68	0	0	0	2.16s
✅ MARKDOWN	markdown-table-formatter	68	0	0	0	1.18s
✅ PYTHON	black	1040	43	0	0	67.12s
✅ PYTHON	isort	1040	64	0	0	4.74s
✅ REPOSITORY	checkov	yes		no	no	51.23s
✅ REPOSITORY	gitleaks	yes		no	no	6.87s
✅ REPOSITORY	git_diff	yes		no	no	0.1s
✅ REPOSITORY	secretlint	yes		no	no	28.36s
✅ REPOSITORY	syft	yes		no	no	3.19s
✅ REPOSITORY	trivy-sbom	yes		no	no	9.12s
✅ REPOSITORY	trufflehog	yes		no	no	25.53s
✅ YAML	prettier	11	0	0	0	0.72s
✅ YAML	v8r	11		0	0	11.21s
✅ 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@9.5.0 --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_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Closing as superseded by the newer active multinomial probability-validation work in #3220, which covers JAX and PyTorch as well.