Fix MegaLinter E402 and invalid parameterized test case causing CI failures

[Copilot]

Two independent CI failures on main: a Ruff E402 lint error in the JAX autodiff test file, and a test collection crash caused by an invalid PartiallyWrappedNormalDistribution instantiation at class-body evaluation time.

Changes

• tests/test_jax_autodiff_backend.py: Add # noqa: E402 to the from pyrecest._backend.jax import autodiff import, which legitimately follows pytest.importorskip() calls that must execute before any JAX import is attempted.

• tests/distributions/test_hypercylindrical_dirac_distribution.py: Remove the invalid "1D Plot" parameterized case. It passed array([1.0]) with bound_dim=1, yielding lin_dim = 1 - 1 = 0, which AbstractLinBoundedCartProdDistribution rejects at construction time — crashing pytest collection before any test ran.

  # Invalid: lin_dim=0 raises ValueError during class body evaluation
  PartiallyWrappedNormalDistribution(
      array([1.0]), array([[1.0]]), bound_dim=1
  )

  The minimum valid distribution requires bound_dim ≥ 1 and lin_dim ≥ 1; the remaining "2D Plot" and "3D Plot" cases already cover the simplest valid configuration.

[github-actions]

Test Results

     8 files  +     8       8 suites  +8   1h 9m 24s ⏱️ + 1h 9m 24s
 1 832 tests + 1 832   1 830 ✅ + 1 830      2 💤 +    2  0 ❌ ±0 
15 064 runs  +15 064  13 632 ✅ +13 632  1 432 💤 +1 432  0 ❌ ±0 

Results for commit 35bb469. ± Comparison against base commit a8ff0ff.

♻️ This comment has been updated with latest results.

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no	no	16.98s
✅ JSON	prettier	2	0	0	0	0.42s
✅ JSON	v8r	2		0	0	2.54s
✅ MARKDOWN	markdownlint	28	0	0	0	1.31s
✅ MARKDOWN	markdown-table-formatter	28	0	0	0	0.36s
✅ PYTHON	bandit	571		0	0	8.85s
✅ PYTHON	black	571	29	0	0	15.89s
✅ PYTHON	flake8	571		0	0	5.3s
✅ PYTHON	isort	571	30	0	0	1.03s
✅ PYTHON	mypy	571		0	0	6.94s
✅ PYTHON	pylint	571		0	0	97.09s
✅ PYTHON	ruff	571	30	0	0	0.12s
✅ REPOSITORY	checkov	yes		no	no	25.78s
✅ REPOSITORY	gitleaks	yes		no	no	13.58s
✅ REPOSITORY	git_diff	yes		no	no	0.08s
✅ REPOSITORY	osv-scanner	yes		no	no	1.02s
✅ REPOSITORY	secretlint	yes		no	no	9.64s
✅ REPOSITORY	syft	yes		no	no	3.38s
✅ REPOSITORY	trivy-sbom	yes		no	no	3.59s
✅ REPOSITORY	trufflehog	yes		no	no	21.06s
✅ YAML	prettier	5	0	0	0	0.5s
✅ YAML	v8r	5		0	0	5.7s
✅ YAML	yamllint	5		0	0	0.46s

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_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_OSV_SCANNER,REPOSITORY_SECRETLINT,REPOSITORY_SYFT,REPOSITORY_TRIVY_SBOM,REPOSITORY_TRUFFLEHOG,YAML_PRETTIER,YAML_YAMLLINT,YAML_V8R

Show us your support by starring ⭐ the repository

[FlorianPfaff]

@copilot CI/CD and megalinter are still failing, fix it

[Copilot]

@copilot CI/CD and megalinter are still failing, fix it

Fixed in commits 738442d and 86b6f5f. The remaining failures were:

MegaLinter:

• Removed unused prod import (test_abstract_hypersphere_subset_distribution.py)
• Removed unused KalmanFilter import (test_multi_bernoulli_mixture_tracker.py)
• Fixed mypy override error: changed convolve signature to accept the supertype HypertoroidalWrappedNormalDistribution (wrapped_normal_distribution.py)
• Fixed pylint E0606 (possibly-used-before-assignment): consolidated import guard with # pylint: disable=possibly-used-before-assignment inline (abstract_hyperspherical_distribution.py)
• Fixed pylint E0102 (function-redefined): renamed duplicate local integrand to integrand_4d (bingham_distribution.py)
• Added pylint disable comments for too-many-arguments, too-many-public-methods, and protected-access

Test failure:

• Fixed test_kernel_sme_filter.py: used float literals in arange() calls (e.g. arange(1.0, 5.0)) so PyTorch produces float tensors instead of Long, which eigvalsh requires

[FlorianPfaff]

@copilot CI/CD is still failing, fix

[Copilot]

@copilot CI/CD is still failing, fix

Pushed additional fixes in commits up to 5e8186e.

MegaLinter is now passing. The test workflow failures are on numpy/Python 3.14 and were pre-existing issues hidden by the collection error. Progress so far on those:

• Fixed Leopardi sampler coordinate ordering: removed erroneous flip(pts_s_leop, axis=0) / flip(grid_eucl, axis=1) and fixed plane-reflection sign from south[:,0] → south[:,-1] — resolves ~7 sampler AssertionErrors
• Added imaginary-part assertion in AbstractOrthogonalBasisDistribution.pdf() sqrt branch — fixes test_pdf_rejects_large_negative_imaginary_part

Remaining failures (~70) are grouped as: 18 spherical-harmonics complex ValueError, "Always using upper hemisphere" grid/filter AssertionErrors across ~30 tests, plus scatter failures in SO3, sine-skewed, Fourier, circular UKF and linear distribution tests — all pre-existing on main, exposed only after the collection error was fixed.