[Gal-Gilor]

Summary

• Renamed Section.section_header, section_text and header_level to header, content and level, matching OutlineNode, and merged MarkdownContent into Section (field order: header, content, level, metadata). to_markdown() and breadcrumb() use the new names.
• SectionMetadata.id is a required int (description kept).
• Section, SectionMetadata and OutlineNode have model_config = ConfigDict(frozen=True). The splitter never assigns attributes after construction, so it needed only the renamed keyword arguments. It does rely on appending to a parent node's children list in get_document_outline, which still works on a frozen model.
• Updated the splitter docstrings and doctests, the tests, the CI wheel check (split_text('# Café')[0].header == 'Café') and the README (field names, examples, Data Models section without MarkdownContent, a sentence on frozen models and model_copy).
• CHANGELOG 0.3.0 "Breaking changes": entries for the renames, the removed MarkdownContent, the required metadata.id and the frozen models. The normalization entry now says anyascii(section.content).
• Renamed test_header_level_counting and test_header_levels_up_to_six to test_level_counting and test_levels_up_to_six, because the issue's grep for header_level must print nothing.

Testing

• uv run ruff check --no-fix .: All checks passed
• uv run ruff format --check .: 9 files already formatted
• uv run pytest -q: 99 passed, including the new test_metadata_requires_id (SectionMetadata(span=(0, 0)) raises ValidationError) and test_models_are_frozen (assigning section.content, section.metadata.id or an outline node's header raises ValidationError; model_copy(update={"content": "x"}) returns a changed copy)
• doctest.testmod on markdown_chunkify.splitter: 13 attempted, 0 failed; on markdown_chunkify.models: 0 attempted, 0 failed
• uv build and the CI wheel check in a fresh venv: passed
• grep -rnE --exclude-dir=__pycache__ "section_header|section_text|header_level|MarkdownContent" markdown_chunkify tests README.md .github: no output