Support negative-kappa Watson Bingham conversion

[FlorianPfaff]

What

• support WatsonDistribution.to_bingham() for negative concentration
• use the Watson axis as the Bingham eigenvector with concentration kappa < 0, with zero concentration on its orthogonal complement
• restore WatsonDistribution.sample() for negative-kappa distributions in dimensions that route through the Bingham sampler
• add regression coverage for density equivalence and higher-dimensional sampling

Why

Negative kappa is a valid Watson distribution and is already supported by the density and mode semantics. However, sample() uses the Bingham representation whenever dim != 2, while to_bingham() previously raised NotImplementedError for every negative kappa. This made the same valid distribution sampleable on S^2 but unusable through the public sampling API on S^1/S^3/etc.

For kappa < 0, the exact Bingham exponent is obtained directly with concentrations [kappa, 0, ..., 0] and the Watson axis as the first eigenvector:

x^T M diag(Z) M^T x = kappa * (mu^T x)^2.

Regression coverage

• negative-kappa Watson and converted Bingham PDFs agree on S^3
• converted concentrations/eigenvector encode the expected axis
• negative-kappa sampling above S^2 returns unit-norm samples instead of raising NotImplementedError

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Max errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no		no	7.63s
✅ JSON	prettier	7	0	0		0	0.79s
✅ JSON	v8r	7		0		0	4.46s
✅ MARKDOWN	markdownlint	68	0	0		0	1.88s
✅ MARKDOWN	markdown-table-formatter	68	0	0		0	0.4s
✅ PYTHON	black	1989	15	0		0	66.63s
✅ PYTHON	isort	1989	27	0		0	3.0s
✅ REPOSITORY	betterleaks	yes		no		no	1.24s
✅ REPOSITORY	checkov	yes		no		no	36.77s
✅ REPOSITORY	git_diff	yes		no		no	0.17s
✅ REPOSITORY	secretlint	yes		no		no	104.87s
✅ REPOSITORY	syft	yes		no		no	3.28s
✅ REPOSITORY	trivy-sbom	yes		no		no	2.23s
✅ YAML	prettier	11	0	0		0	0.56s
✅ YAML	v8r	11		0		0	19.07s
✅ YAML	yamllint	11		0		0	1.12s

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