Accept scalar and array-like backend products

[FlorianPfaff]

Pull request

Summary

• handle scalar operands in shared backend dot helpers by multiplying instead of using vector einsum paths
• coerce array-like inputs for backend outer and matvec helpers
• add backend contract coverage for scalar dot/outer and array-like dot/outer/matvec inputs
• fix PyTorch dot scalar handling (0-D operands) to use multiplication before einsum/dot paths, resolving the CI failures in test_dot_accepts_scalar_operands

Checklist

• Tests were added or updated.
• Backend limitations were documented or added to src/pyrecest/_backend/capabilities.py.
• Shape, dtype, and measurement-set conventions were checked against docs/conventions.md.
• New user-facing failures use clear messages or shared exceptions from pyrecest.exceptions.
• Public examples and docs were updated when relevant.
• Performance-sensitive changes were checked against benchmarks/basic_regressions.py or a focused benchmark.
• python scripts/check_release_consistency.py --local-only passes when release metadata changed.
• The package smoke path still works: python -m build, twine check dist/*, install wheel, run a basic example.

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no	no	56.12s
✅ JSON	prettier	7	0	0	0	0.9s
✅ JSON	v8r	7		0	0	5.31s
✅ MARKDOWN	markdownlint	68	0	0	0	1.89s
✅ MARKDOWN	markdown-table-formatter	68	0	0	0	0.67s
✅ PYTHON	black	823	0	0	0	51.59s
✅ PYTHON	isort	823	0	0	0	3.61s
✅ REPOSITORY	checkov	yes		no	no	45.93s
✅ REPOSITORY	gitleaks	yes		no	no	12.83s
✅ REPOSITORY	git_diff	yes		no	no	0.13s
✅ REPOSITORY	secretlint	yes		no	no	25.26s
✅ REPOSITORY	syft	yes		no	no	5.41s
✅ REPOSITORY	trivy-sbom	yes		no	no	7.67s
✅ REPOSITORY	trufflehog	yes		no	no	25.39s
✅ YAML	prettier	11	0	0	0	0.79s
✅ YAML	v8r	11		0	0	10.69s
✅ YAML	yamllint	11		0	0	0.43s

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

[FlorianPfaff]

@copilot Fix the CI/CD tests

[Copilot]

@copilot Fix the CI/CD tests

Fixed in ca3eacb: I updated the PyTorch backend dot implementation to handle scalar operands via _torch.multiply before the einsum/dot paths, which addresses the failing CI tests (test_dot_accepts_scalar_operands).