Port GSSVMDistribution and fix bessel_ratio normalization bug

[Copilot]

Adds GSSVMDistribution (Generalized Skew-Symmetric Von Mises) to match the libDirectional API, and fixes a pre-existing bug in bessel_ratio that made the PDF produce NaN for all orders m > 1.

Bug fix: bessel_ratio

The function was computing I_p(z) / I_p(0), but I_p(0) = 0 for all p > 0, causing division by zero. The normalization constants for m=2,3,4 require I_p(κ) / I_0(κ):

# Before (always NaN/inf for p > 0)
return ive(p, z) / ive(p, 0)

# After
return ive(p, z) / ive(0, z)

New class: GSSVMDistribution

• Subclass of GeneralizedKSineSkewedVonMisesDistribution with k=1 fixed
• Exposes n (order) instead of m, matching libDirectional naming
• shift() returns a GSSVMDistribution rather than the base type
• Exported from pyrecest/distributions/__init__.py

dist = GSSVMDistribution(mu=0.5, kappa=2.0, lambda_=0.3, n=2)
dist.pdf(xs)      # works correctly for all supported orders 1–4
dist.shift(pi/4)  # returns GSSVMDistribution

Tests

New test file test_gssvm_distribution.py covers initialization, parameter validation, n↔m mapping, non-negative PDFs for n=1..4, PDF parity with the parent class, shift() type/value, mu wrapping, and NotImplementedError for n > 4.

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no	no	9.33s
✅ JSON	prettier	2	0	0	0	0.56s
✅ JSON	v8r	2		0	0	2.44s
✅ MARKDOWN	markdownlint	1	0	0	0	0.68s
✅ MARKDOWN	markdown-table-formatter	1	0	0	0	0.23s
✅ PYTHON	bandit	251		0	0	3.14s
✅ PYTHON	black	251	1	0	0	4.44s
✅ PYTHON	flake8	251		0	0	1.77s
✅ PYTHON	isort	251	1	0	0	0.49s
✅ PYTHON	mypy	251		0	0	4.16s
✅ PYTHON	pylint	251		0	0	71.78s
✅ PYTHON	ruff	251	1	0	0	0.05s
✅ REPOSITORY	checkov	yes		no	no	23.93s
✅ REPOSITORY	gitleaks	yes		no	no	4.45s
✅ REPOSITORY	git_diff	yes		no	no	0.02s
✅ REPOSITORY	secretlint	yes		no	no	5.89s
✅ REPOSITORY	syft	yes		no	no	3.35s
✅ REPOSITORY	trivy-sbom	yes		no	no	2.53s
✅ REPOSITORY	trufflehog	yes		no	no	16.52s
✅ YAML	prettier	4	0	0	0	0.5s
✅ YAML	v8r	4		0	0	5.03s
✅ YAML	yamllint	4		0	0	0.44s

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@9.4.0 --custom-flavor-setup --custom-flavor-linters PYTHON_PYLINT,PYTHON_BLACK,PYTHON_FLAKE8,PYTHON_ISORT,PYTHON_BANDIT,PYTHON_MYPY,PYTHON_RUFF,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

[github-actions]

Test Results

   12 files  ±  0     12 suites  ±0   3h 18m 24s ⏱️ + 4m 29s
  606 tests +  9    606 ✅ +  9      0 💤 ±0  0 ❌ ±0 
7 484 runs  +108  6 236 ✅ +108  1 248 💤 ±0  0 ❌ ±0 

Results for commit 7670a61. ± Comparison against base commit 96e327d.

♻️ This comment has been updated with latest results.

[FlorianPfaff]

@copilot Somewhere, pyrecest.backend is not used and some numpy object is created although it shouldn't be.
=========================== short test summary info ============================
FAILED pyrecest/tests/distributions/test_gssvm_distribution.py::TestGSSVMDistribution::test_pdf_matches_generalized_parent - TypeError: unsupported operand type(s) for *: 'numpy.ndarray' and 'Tensor'
FAILED pyrecest/tests/distributions/test_gssvm_distribution.py::TestGSSVMDistribution::test_pdf_non_negative - TypeError: unsupported operand type(s) for *: 'numpy.ndarray' and 'Tensor'
= 2 failed, 512 passed, 66 skipped, 5559611 warnings, 11 subtests passed in 801.23s (0:13:21) =

[FlorianPfaff]

@copilot Somewhere, pyrecest.backend is not used and some numpy object is created although it shouldn't be.
=========================== short test summary info ============================
FAILED pyrecest/tests/distributions/test_gssvm_distribution.py::TestGSSVMDistribution::test_pdf_matches_generalized_parent - TypeError: unsupported operand type(s) for *: 'numpy.ndarray' and 'Tensor'
FAILED pyrecest/tests/distributions/test_gssvm_distribution.py::TestGSSVMDistribution::test_pdf_non_negative - TypeError: unsupported operand type(s) for *: 'numpy.ndarray' and 'Tensor'
= 2 failed, 512 passed, 66 skipped, 5559611 warnings, 11 subtests passed in 801.23s (0:13:21) =

[Copilot]

@copilot Somewhere, pyrecest.backend is not used and some numpy object is created although it shouldn't be.
=========================== short test summary info ============================
FAILED py...

Fixed in 65cb132. scipy.stats.vonmises.pdf was the culprit — it always returns a NumPy array regardless of the active backend. Replaced it with VonMisesDistribution(self.mu, self.kappa).pdf(xs), which uses the pyrecest backend operations (exp, cos) and returns the correct tensor type.