Improve landing page SEO and AI discovery

[aryamantodkar]

Summary

Make the Premove landing page discoverable at premove.dev and set its browser title to Premove: Open Source Voice Infra.

Why

The live landing page currently has no robots.txt, sitemap.xml, or llms.txt. The same content also appears at the Mintlify docs root. Search engines and AI search crawlers need a clear canonical home and a crawlable page.

Changes

• Add a one-page sitemap and a robots file that allows general crawlers and OAI-SearchBot and advertises the sitemap.
• Add an llms.txt summary linking the landing page, Premove ITN, and the live Mintlify AI index.
• Add Open Graph metadata and WebSite/Organization JSON-LD to the landing page.
• Set the browser and preview title to Premove: Open Source Voice Infra. Keep the visible H1 unchanged.
• Mark only the duplicate Mintlify root page noindex. The ITN docs routes keep their own canonical URLs.
• Add a focused test for title, canonical, structured data, sitemap, robots access, and AI index links.

Validation

• uv run --locked python scripts/agent/check.py focused --pytest tests/test_docs_routes.py passed.
• git diff --check passed.
• Local HTTP checks served /, /robots.txt, /sitemap.xml, and /llms.txt successfully.
• Verified the live https://docs.premove.dev/llms.txt responds with HTTP 200.

Deployment

The current Vercel project deploys from the personal fork aryamantodkar/premove-itn. After merging this PR, sync that fork's main branch and redeploy. Then verify the three new files on premove.dev and submit https://premove.dev/sitemap.xml in Google Search Console.

An llms.txt file does not guarantee AI search inclusion or improve Google rankings. Google states that its Search systems ignore llms.txt; crawl access, canonical URLs, and useful content are the primary signals.