feat: Decouple YouTube analytics, add syndication schema references, and wire syndication credentials - #706

Summary of Changes

1. Decoupled YouTube Analytics:

   • Added scheduled function sync-youtube-analytics running daily at 6:00 AM UTC.
   • Uses YouTube Data API v3 to batch up to 50 videos per request, fetching views/likes/comments without re-scraping captions.
   • Updated sync-youtube-transcript to never re-scrape captions if a transcript is already marked completed.
   • Patches statistics.youtube on source documents and statistics on the referenced transcript document.

2. Syndication Schema References & Updates:

   • Added syndications reference array to apps/sanity/schemas/partials/content.ts pointing to syndication documents.
   • Maintained devto and hashnode fields for backwards compatibility.
   • Updated syndicate-content to link created/updated syndication documents into the parent syndications array.
   • Updated Hashnode GraphQL endpoint to https://gql-beta.hashnode.com.
   • Wired PRIVATE_DEVTO and PRIVATE_HASHNODE secrets through deploy.yml and sanity.blueprint.ts.

3. Content Backfill:

   • Created scripts/backfill-content-from-transcripts.mjs and workflow .github/workflows/backfill-content.yml to synthesize and backfill rich Portable Text content for episodes with completed transcripts and empty content.