fix(markdown): parse escaped wikilink aliases

[zagushka]

What does this PR do?

Obsidian requires an escaped alias separator inside Markdown tables:

[[notes/Target\|Alias]]

The Markdown wikilink regex currently consumes the backslash as part of the target. The extractor then looks for notes/Target\.md, so it cannot stamp the real target_file. This can leave the direct and incremental extraction paths without the canonical target path.

This change accepts both | and \| as alias separators. It adds a regression test that requires both forms to resolve to the same existing Markdown file.

Related to #295. This focused parser fix does not close that broader structural-extraction issue.

Type of change

• Bug fix
• New feature
• Documentation
• Tests or CI
• Refactor
• Security fix

How was this tested?

uv run --frozen pytest tests/ -q --tb=short
5937 passed, 14 skipped

uv run --frozen pytest tests/test_languages.py -q
396 passed, 20 skipped

uv run --frozen ruff check graphify/extractors/markdown.py tests/test_languages.py
All checks passed!

Graphify-specific checklist

• I added or updated tests for behavior changes.
• I updated documentation or confirmed that no documentation is needed.
• I updated generated skill artifacts when changing their source fragments. (Not applicable: no skill source changed.)
• I considered compatibility across supported Python versions.
• I confirmed that no API keys, generated graph data, or local-only files are included.