Fix nested matrix cached data preparation

[FlorianPfaff]

Summary

• keep the nested-matrix MEG data preparation action active even when use_nextcloud_data=false
• interpret use_nextcloud_data as permission to download from WebDAV on cache miss, not as permission to skip cached/self-hosted data exposure
• remove the unused USE_NEXTCLOUD_DATA job environment variable and pass the workflow input to download-on-miss

Context

PR #155 added the manual sharded nested stimulus matrix workflow. In the merged workflow, setting use_nextcloud_data=false skipped .github/actions/prepare-meg-data entirely. That also skipped exposing an existing self-hosted/local cache at .cache/meg-data-main, so the later Validate main data files step would fail even when the data were already present in the runner cache.

Validation

• Inspected the workflow through the GitHub connector.
• Compared the branch against main; only this workflow file changed (+2/-3).
• Local workflow execution was not possible in this environment because it cannot resolve github.com for a clone/install.

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no	no	3.13s
✅ MARKDOWN	markdownlint	13	0	0	0	0.88s
✅ PYTHON	ruff	126	0	0	0	0.3s
✅ REPOSITORY	git_diff	yes		no	no	0.02s
✅ YAML	prettier	6	0	0	0	0.46s
✅ YAML	v8r	6		0	0	3.6s
✅ YAML	yamllint	6		0	0	0.49s

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

Show us your support by starring ⭐ the repository