[codex] Add LEAP Finetune docs

[alay2shah]

Summary

• Add a new LEAP Finetune guide under lfm/fine-tuning and wire it into navigation, redirects, homepage links, and related fine-tuning pages.
• Add model cards and listings for LFM2.5-VL-1.6B-Extract, LFM2.5-VL-450M-Extract, and LFM2.5-Audio-1.5B-JP.
• Update the complete model library, link snapshot, and maintainer model matrix for the new pages.

Validation

• node -e "JSON.parse(require('fs').readFileSync('docs.json','utf8')); console.log('docs.json ok')"
• git diff --cached --check
• npm run snapshot:check
• Local route smoke check for the three new model pages returned 200.

Resolve https://linear.app/liquid-ai/issue/DOC-61/request-leap-finetune-does-not-appear-under-fine-tunning.

[mintlify]

Preview deployment for your docs. Learn more about Mintlify Previews.

Project	Status	Preview	Updated (UTC)
liquid-docs	🟢 Ready	View Preview	Jun 12, 2026, 3:52 PM

💡 Tip: Enable Workflows to automatically generate PRs for you.

[mlabonne]

Can we add LFM2.5-Embedding-350M and ColBERT to Liquid Nanos?

[alay2shah]

Can I also deprecate the 2 text extracts, RAG, and math (not sure if 2.5 350M beats 350M-Math)?

[mlabonne]

Let's deprecate LFM2-1.2B-Extract, LFM2-350M-Extract, and LFM2-1.2B-RAG. I wouldn't deprecate the Math one because it's still really good on math benchmarks.