Fix PyTorch std array-like backend inputs

[FlorianPfaff]

Summary

• Wrap the public PyTorch backend std export so NumPy-style array-like inputs are coerced before calling torch.std.
• Cast integer inputs to the configured default floating dtype, matching NumPy/JAX-style std behavior rather than leaking PyTorch's floating/complex-only requirement.
• Add a backend-portability regression test that forces PYRECEST_BACKEND=pytorch.

Rationale

pyrecest.backend.std([[1, 2, 3], [4, 5, 6]], axis=0, ddof=1, keepdims=True) should work like the NumPy/JAX backends. Before this change, the PyTorch backend path could forward list or integer tensor inputs to torch.std, which rejects non-tensor list inputs and integer tensors.

Validation

• Connector compare confirms the branch is 2 commits ahead of current main and 0 behind.
• Added tests/test_backend_std_contract.py for active-backend and forced-PyTorch coverage.
• Full local pytest could not be run because the execution container cannot resolve github.com; PR CI should run the focused regression and full matrix.

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no	no	52.57s
✅ JSON	prettier	7	0	0	0	1.14s
✅ JSON	v8r	7		0	0	4.63s
✅ MARKDOWN	markdownlint	66	0	0	0	2.2s
✅ MARKDOWN	markdown-table-formatter	66	0	0	0	0.84s
✅ PYTHON	black	763	0	0	0	40.34s
✅ PYTHON	isort	763	1	0	0	2.17s
✅ REPOSITORY	checkov	yes		no	no	46.27s
✅ REPOSITORY	gitleaks	yes		no	no	16.29s
✅ REPOSITORY	git_diff	yes		no	no	0.11s
✅ REPOSITORY	secretlint	yes		no	no	29.82s
✅ REPOSITORY	syft	yes		no	no	3.77s
✅ REPOSITORY	trivy-sbom	yes		no	no	4.52s
✅ REPOSITORY	trufflehog	yes		no	no	22.3s
✅ YAML	prettier	11	0	0	0	0.69s
✅ YAML	v8r	11		0	0	12.7s
✅ YAML	yamllint	11		0	0	0.59s

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_BLACK,PYTHON_ISORT,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