feat: extract markdown for PDF pages with password

[aynugek]

Summary by cubic

Add password support to per-page Markdown extraction so encrypted PDFs can be processed. Previously, extract_pages_markdown failed on encrypted PDFs; now callers can pass a password or rely on an empty-password attempt. Wrong or missing passwords return an Encrypted error.

• API changes

  • Rust: Add extract_pages_markdown_with_password(path, pages, password) and extract_pages_markdown_mem_with_password(bytes, pages, password). Existing functions remain and delegate with None.
  • Python: extract_pages_markdown(path, pages=None, password=None) and extract_pages_markdown_bytes(data, pages=None, password=None). Stubs and docs updated.
  • Node (N-API): extract_pages_markdown(buffer, pages?, password?) and extract_pages_markdown_async(buffer, pages?, password?). Optional password; default tries empty password. Type definitions updated. No migration required; existing two-arg calls continue to work.

• Notes

  • Error behavior: encrypted files without a correct password return PdfError::Encrypted (propagates to Python/Node).
  • Docs: Python and Rust API guides updated; new integration test covers decryption and error cases.

^{Written for commit 9b1ccac. Summary will update on new commits.}

[cubic-dev-ai]

No issues found across 7 files

<sub>Shadow auto-approve: would auto-approve. Adds optional password support for extract_pages_markdown in Rust core and Python/Node bindings with a new integration test. Backward compatible, documented, and tested; no risky changes are visible, so auto-approval.

Re-trigger cubic</sub>