Filter pale chapters to save only chapter content

[Tomotz]

Filter pale chapters to save only chapter content

Summary

Adds extract_chapter_content() to save_ebook.py that uses regex to extract only the chapter title (<h1 class="entry-title">) and body (<div class="entry-content">) from each downloaded WordPress page. This strips out user comments, sharing/like buttons (jp-post-flair), related posts (jp-relatedposts), navigation, sidebar, footer metadata, and all other non-chapter HTML.

The function is applied in the pale download loop so pale_full.html contains only chapter text.

Review & Testing Checklist for Human

• Regex vs nested divs: The content regex relies on the </div><!-- .entry-content --> HTML comment as its end marker. Verify this comment is consistently present across all 311 chapters — if any chapter's template differs, content could be truncated or missed entirely.
• Silent chapter drops: If a page returns unexpected HTML (error page, changed template), both regexes return empty strings and the chapter is silently skipped. Consider whether a warning/print should be added.
• jp-post-flair stripping: The cleanup regexes use greedy .* with re.DOTALL, removing everything from those divs to end-of-string. Confirm these divs always appear at the tail of entry-content and never inside actual chapter text.
• Test run: Re-run the script on a handful of chapters (or the full set) and spot-check that output contains complete chapter text with no leftover comment/sharing HTML and no missing content.

Notes

• Link to Devin run: https://app.devin.ai/sessions/9c54fafced7e4cd1b667439e04eb40b1
• Requested by: @Tomotz