Add CVFileLoader integration page

[Ilansky (ilanoh)]

This PR adds a document loader integration page for langchain-cvfile, which loads .cv files (an open file format for resumes/CVs: PDF/A-3u with embedded Markdown/HTML/JSON payloads carried as PDF Associated Files).

Why

A .cv file is a PDF/A-3u that ships a Markdown copy of the same content (plus optional HTML and JSON Resume) as PDF Associated Files (ISO 32000-2 §14.13). For RAG/ATS use cases, the embedded Markdown is a much cleaner text representation than OCRing the visual PDF layer. This loader pulls those embedded payloads directly and returns one Document per textual payload, with the payload declared as cv:primaryPayload in the file's XMP metadata flagged metadata["primary"] = True.

What this PR contains

A single new MDX page at src/oss/python/integrations/document_loaders/cvfile.mdx, following the existing PyPDFLoader / similar loader template (frontmatter, integration details table, loader features table, setup, initialization, load + lazy_load examples, metadata reference, API reference link).

Package details

• PyPI: https://pypi.org/project/langchain-cvfile/ (v0.1.0 live, Apache-2.0)
• Source: https://github.com/cvfile/cv/tree/main/integrations/langchain-cvfile
• Spec: https://cvfile.org/spec/
• Dependencies: cvfile>=0.1.0,<1, langchain-core>=0.3,<1

Code sample is runnable

The example in the page works against any .cv file produced by the reference SDK. Sample output in the page was generated against an actual fixture; running loader.load() returns three Documents for a typical resume (resume.md primary, resume.html alternate, resume.json alternate).

[github-actions]

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

• @mdrxy (Python integrations)