Stabilize concentrated Watson density evaluation

[FlorianPfaff]

Bug

WatsonDistribution.pdf(...) evaluated the density as the product of the normalization constant and an exponential term.

For large positive concentration, those two factors can underflow and overflow separately even though the normalized density is finite. With mu=[1, 0, 0], kappa=1000, and evaluation at the mode, the normalization constant underflows to zero while exp(kappa) overflows, producing 0 * inf -> NaN. The mathematically correct density is approximately 159.075285843.

Fix

Evaluate the same density in log space using the class's existing stable ln_norm_const:

exp(ln_norm_const + kappa * (x @ mu) ** 2)

This preserves ordinary-case values while avoiding the spurious 0 * inf intermediate.

Regression coverage

Add a focused regression at kappa=1000 that evaluates the mode and checks the finite expected density.

Validation

• independently reproduced the previous underflow/overflow pair and resulting NaN;
• independently evaluated the log-space expression and obtained 159.07528584318962;
• final diff is limited to one production-line replacement and six regression-test lines;
• branch is based directly on current main (f196ec41d1d8681e279936db9f1a0a3abfa276e7) and is 2 commits ahead / 0 behind;
• GitHub Actions is authoritative for the full backend, lint, packaging, documentation, and integration matrix.

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no	no	20.33s
✅ JSON	prettier	7	0	0	0	0.76s
✅ JSON	v8r	7		0	0	6.63s
✅ MARKDOWN	markdownlint	68	0	0	0	1.45s
✅ MARKDOWN	markdown-table-formatter	68	0	0	0	0.6s
✅ PYTHON	black	1866	87	0	0	69.18s
✅ PYTHON	isort	1866	140	0	0	2.1s
✅ REPOSITORY	betterleaks	yes		no	no	1.98s
✅ REPOSITORY	checkov	yes		no	no	39.25s
✅ REPOSITORY	gitleaks	yes		no	no	13.05s
✅ REPOSITORY	git_diff	yes		no	no	0.27s
✅ REPOSITORY	secretlint	yes		no	no	61.63s
✅ REPOSITORY	syft	yes		no	no	4.69s
✅ REPOSITORY	trivy-sbom	yes		no	no	4.49s
✅ REPOSITORY	trufflehog	yes		no	no	24.83s
✅ YAML	prettier	11	0	0	0	0.49s
✅ YAML	v8r	11		0	0	8.52s
✅ YAML	yamllint	11		0	0	0.47s

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