feat: add documentation crawler for OpenZeppelin docs

[enitrat]

PR Summary: feat: add documentation crawler for OpenZeppelin docs

📜 High-Level Summary

This PR replaces the AsciiDoc-based OpenZeppelin documentation ingestion with a simpler approach using pre-crawled markdown files. The crawler fetches documentation from websites using sitemaps and converts HTML content to clean markdown, making it easier to maintain up-to-date documentation sources for the Cairo Coder RAG pipeline.

Changeset 1: Replace AsciiDoc with Pre-Crawled Markdown for OpenZeppelin Docs

Files Affected:

• packages/ingester/src/ingesters/OpenZeppelinDocsIngester.ts
• packages/ingester/src/utils/RecursiveMarkdownSplitter.ts
• packages/ingester/asciidoc/oz-playbook.yml (deleted)
• packages/ingester/asciidoc/playbook.yml

Summary of Changes:

• Modified OpenZeppelinDocsIngester to read from a pre-crawled markdown file instead of processing AsciiDoc
• Added RecursiveMarkdownSplitter utility with comprehensive markdown-aware chunking logic
• Removed the separate oz-playbook.yml configuration as it's no longer needed
• Updated imports and chunk processing to use the new markdown splitter

[TRIAGE]: NEEDS_REVIEW

---

Changeset 3: Add Pre-Crawled OpenZeppelin Documentation

Files Affected:

• python/scripts/summarizer/generated/openzeppelin_docs_summary.md

Summary of Changes:

• Added a pre-crawled snapshot of the OpenZeppelin documentation dated 2025-08-02
• This file serves as the source for the OpenZeppelinDocsIngester

[TRIAGE]: NEEDS_REVIEW

[ijusttookadnatest]

Great PR, everything's ok