Clean public repository content

[KaiCao2003]

Summary

• replace the README and public docs with concise project, setup, and architecture guidance
• remove tracked VesSAP/LAMBADA NPZ data, manifests, the copied data license, and unused LAMBADA code/tests/scripts
• load the optional VesSAP package from the user data directory and advertise the capability only when the files are present
• remove internal audit/status artifacts and ignore PDFs and scientific data archives

Verification

• parsed all six changed Python files
• scanned the complete remote tree: no tracked PDF/NPZ/NPY/NIfTI/7z files
• scanned repository text: no targeted disclaimer wording, LAMBADA references, or deleted-path references
• checked all relative Markdown links
• GitHub Actions pending