feat: add EPUB book Markdown serializer

[salmonumbrella]

What changed

Adds EpubDocSerializer, a MarkdownDocSerializer variant that renders a book as Markdown preceded by a YAML frontmatter block holding the book metadata and a chapter index, plus DoclingDocument.export_to_epub() / save_as_epub().

Every chapter entry carries the absolute UTF-8 byte offset and the 1-based line number of its heading, so a reader can seek straight to a chapter instead of unpacking the source or scanning the whole book.

---
title: "Alice's Adventures in Wonderland"
authors: ["Lewis Carroll"]
language: "en"
chapters:
  - title: "CHAPTER I. Down the Rabbit-Hole"
    line:         12
    byte:        287
---

Why

Requested by @dolfim-ibm in review of docling-project/docling#3993, which had the same serializer living in docling/backend/epub_serializer.py:

The serializer should be in docling-core, and I would propose to add Epub as a standalone output format

The companion docling PR is being restructured to drop its local serializer and consume this one behind a new epub output format.

Notes for reviewers

Two details are load-bearing and have dedicated tests:

• The offsets are advertised inside the block they measure. The frontmatter is rendered twice: once to measure how far it shifts the body, once with the shifted offsets. Padding every offset to a fixed width keeps the two renderings byte-identical; a guard raises if that ever stops holding.
• A chapter is a TitleItem or a level-1 SectionHeaderItem. An EPUB spine document may title its chapters with either <h1> or <h2>, and the HTML backend maps <h1> to a title item. Matching only section headers leaves the index empty for every book that uses <h1> per chapter.

save_as_epub() writes raw UTF-8 bytes rather than going through text mode, since the advertised offsets index the saved file and Windows newline translation would invalidate them.

Checklist

• Tests added (test/test_epub_serialization.py, 9 cases)
• pytest test — 606 passed, 6 skipped
• ruff check / ruff format / mypy clean
• Generated docs regenerated: no schema drift

🤖 Generated with Claude Code

[github-actions]

✅ DCO Check Passed

Thanks @salmonumbrella, all your commits are properly signed off. 🎉

[mergify]

Merge Protections

🔴 1 of 2 protections blocking · waiting on 👀 reviews

	Protection	Waiting on
🔴	Require two reviewer for test updates	👀 reviews
🟢	Enforce conventional commit	—

🔴 Require two reviewer for test updates

Waiting for

• #approved-reviews-by >= 2

This rule is failing.

When test data is updated, we require two reviewers

• #approved-reviews-by >= 2

---

Show 1 satisfied protection

🟢 Enforce conventional commit

Make sure that we follow https://www.conventionalcommits.org/en/v1.0.0/

• title ~= ^(fix|feat|docs|style|refactor|perf|test|build|ci|chore|revert)(?:\(.+\))?(!)?:

[codecov]

Codecov Report

❌ Patch coverage is 95.50562% with 4 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
docling_core/transforms/serializer/epub.py	97.33%	2 Missing ⚠️
docling_core/types/doc/document.py	85.71%	2 Missing ⚠️

📢 Thoughts on this report? Let us know!