Fix partially wrapped-normal input edge cases

[FlorianPfaff]

Summary

• coerce partially wrapped-normal mu, C, query points, and set_mean() inputs to backend arrays before inspecting shapes
• normalize 1-D query-point inputs consistently, matching the wrapped-normal/hypertoroidal behavior for scalar domains
• handle bound_dim == 0 without constructing an invalid empty meshgrid
• fix linear_mean() for fully periodic distributions; self.mu[-0:] previously returned the full state instead of an empty linear component
• add regression tests for Python sequence inputs, scalar-domain query batches, set_mean() wrapping, and fully periodic linear-component access

Validation

• connector compare: branch is 3 commits ahead of main, 0 behind
• changed files are limited to partially_wrapped_normal_distribution.py and its focused tests

Note: I could not run the full test suite locally in this sandbox because direct GitHub checkout is unavailable here; PR CI should run the suite.

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no	no	33.23s
✅ PYTHON	black	4	1	0	0	2.02s
✅ PYTHON	isort	4	1	0	0	0.55s
✅ REPOSITORY	checkov	yes		no	no	30.64s
✅ REPOSITORY	gitleaks	yes		no	no	15.68s
✅ REPOSITORY	git_diff	yes		no	no	0.05s
✅ REPOSITORY	secretlint	yes		no	no	15.35s
✅ REPOSITORY	syft	yes		no	no	2.19s
✅ REPOSITORY	trivy-sbom	yes		no	no	3.85s
✅ REPOSITORY	trufflehog	yes		no	no	22.67s

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
Set VALIDATE_ALL_CODEBASE: true in mega-linter.yml to validate all sources, not only the diff

Show us your support by starring ⭐ the repository

[github-actions]

Test Results

0 tests   0 ✅  0s ⏱️
0 suites  0 💤
0 files    0 ❌

Results for commit 6c7c445.