Split binary into chunks if needed

[tan-z-tan]

WHY

Firestore has a strict document size limit of 1MB. When the Yjs document state exceeds this limit, write operations fail, preventing users from saving large documents. We need a mechanism to handle documents larger than 1MB.

WHAT

• Implemented Chunking for Large Documents:
  • Modified saveToFirestore to check if the encoded Yjs document size exceeds 800KB.
  • If the size limit is exceeded, the document is split into multiple chunks.
  • These chunks are saved as individual documents in a yfire_chunks subcollection under the main document.
  • The main document is updated with a flag chunked: true and the content field is cleared.
• Updated Loading Logic:
  • Modified trackData to handle chunked: true documents.
  • It now fetches all chunks from the yfire_chunks subcollection, reassembles them in order, and applies the update to the Yjs document.
• Removed GCS Logic:
  • Removed the incomplete and unused Google Cloud Storage (GCS) integration code to keep the provider focused and clean.

[tan-z-tan]

@deathg0d Hi.
This PR was originally intended for my fork, but if you find it useful, I’d be happy if you consider merging it. If not, feel free to close it.