Parallelize markdown generation with ProcessPoolExecutor

[svekars]

• Parallelize HTML-to-markdown conversion using ProcessPoolExecutor for significant speedup on large doc sites (69x
  faster on 2,563 PyTorch Python doc pages)
• Refactor _generate_md_files to return a dict of {docname: md_content} instead of a count, passing in-memory content
  to _generate_llms_full_txt to avoid redundant disk reads
• Update tests to match the new API and add parallelism-specific tests validated against real PyTorch docs

pytorch_sphinx_theme2/llm_generation.py:

• Extract _convert_single_doc() as a top-level function (required for ProcessPoolExecutor pickling)
• _generate_md_files() now uses ProcessPoolExecutor to convert docs in parallel; returns dict[docname, md_content]
  instead of int
• _generate_llms_full_txt() takes the in-memory md_contents dict instead of re-reading .md files from disk

tests/test_llm_markdown.py:

• Fix 7 tests (TestGenerateMdFiles, TestGenerateLlmsFullTxt) to match new return types and signatures
• Add TestParallelConversion class with 4 tests validating correctness and performance

Benchmarks

Dataset	Pages	Sequential	Parallel	Speedup
PyTorch Python docs	2,563	622.5s (~10 min)	9.0s	69.4x

All parallel results match sequential output byte-for-byte.

Test plan

• All 37 unit tests pass (pytest tests/test_llm_markdown.py)
• Parallel output matches sequential output byte-for-byte on 66 PyTorch C++ doc pages
• Parallel output matches sequential output byte-for-byte on 2,563 PyTorch Python doc pages
• Verified .md files are written to disk correctly
• 200-file synthetic stress test passes

Tested against pytorch docs pytorch/pytorch#179268

[netlify]

✅ Deploy Preview for pytorchsphinxtheme ready!

Name	Link
🔨 Latest commit	c222c62
🔍 Latest deploy log	https://app.netlify.com/projects/pytorchsphinxtheme/deploys/69d42613605a2b0008a81499
😎 Deploy Preview	https://deploy-preview-243--pytorchsphinxtheme.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify project configuration.

[AlannaBurke]

LGTM!