feat(bookstack): use native markdown export for pages

[hexsm]

Summary

Switches the BookStack connector from plain text output to BookStack's native Markdown page export.

BookStack synced pages now use .md files backed by:

GET /api/pages/{id}/export/markdown

Why

The connector previously fetched page HTML and stripped tags with a simple regex before uploading .txt files. This lost useful document structure such as headings, lists, links, code blocks, and tables.

BookStack provides a native Markdown export endpoint for pages, which preserves more structure and is better suited for Knowledge Base ingestion.

Changes

• Use .md filenames for BookStack synced pages.
• Fetch page content from BookStack's native Markdown export endpoint.
• Remove the manual HTML-to-text regex conversion.
• Document the behavior change in CHANGELOG.md.

Validation

• python3 -m compileall src/oikb/connectors/bookstack.py
• Tested against a running BookStack/oikb container by patching the connector in-place and confirming Markdown export sync works.

Notes

Existing BookStack .txt entries may be replaced by .md entries on the next sync because the exported file extension changes.

[hexsm]

Closing this in favor of #84.