fix: recognize tabs after Markdown header prefixes

[sunlishuo25]

Related Issues

No existing issue.

Proposed Changes:

MarkdownHeaderSplitter ignores headings such as #\tGuide, leaving their sections together and losing header metadata used by filtering. Accept spaces or tabs after the opening hashes, as allowed by ATX heading syntax.

Add regression coverage for heading hierarchy, both keep_headers modes, secondary splitting, code fences, and synchronous/asynchronous indexing followed by section retrieval. Update the docstring, documentation, and release note.

How did you test it?

The regressions fail on the original source (9 failed, 71 passed). The targeted splitter, converter, writer, and filter retriever unit suites pass after the fix (111 passed, 1 integration test deselected) on Python 3.10. Scoped mypy and pre-commit checks pass.

Tests used a local Hatch environment with runtime dependencies and lightweight test tools. The complete official test environment and CI were not run.

Notes for the reviewer

Developed with GPT assistance.

Checklist

• I have read the contributors guidelines and the code of conduct.
• I have added unit tests and updated the docstrings.
• I've used a conventional commit type for the PR title.
• I have documented my code.
• I have added a release note file.
• I have run pre-commit hooks and fixed any issue.

[vercel]

@sunlishuo25 is attempting to deploy a commit to the deepset Team on Vercel.

A member of the Team first needs to authorize it.