docs: add Mixpeek integration (tools/toolkit + vector store)

[Ethan Steininger (esteininger)]

Summary

Adds documentation pages for the langchain-mixpeek integration package — multimodal search, ingestion, and monitoring for AI agents.

New pages:

• src/oss/python/integrations/tools/mixpeek.mdx — MixpeekTool and MixpeekToolkit (6-tool agent suite)
• src/oss/python/integrations/vectorstores/mixpeek.mdx — MixpeekVectorStore with multimodal ingestion

What Mixpeek adds to the LangChain ecosystem:

• Multimodal retrieval — agents can search video, images, audio, and documents (not just text)
• 6-tool agent toolkit: search, ingest, process, classify, cluster, alert
• VectorStore supporting 6 content types: text, image, video, audio, PDF, Excel
• Bridge methods: as_tool(), as_retriever(), as_toolkit() for interface conversion
• Platform features: taxonomy classification, clustering, alert monitoring

Package details:

• PyPI: langchain-mixpeek (v0.4.3, 89 tests passing)
• npm: @mixpeek/langchain (JS companion)
• GitHub: mixpeek/langchain-mixpeek
• License: MIT
• PEP 561 compliant (py.typed)

Both pages follow the official templates and include setup, instantiation, invocation, and agent usage examples.

[github-actions]

Thanks for opening a docs PR, Ethan Steininger (@esteininger)! When it's ready for review, please add the relevant reviewers:

• @mdrxy (Python integrations)