* feat: register open ai token usage during indexing
* feat: register open ai token usage during search
* chore: fixed open ai token usage when searching for documents
* feat: implement appflowy-indexer and search api endpoint
* chore: store embedding content type
* chore: more descriptive messages for open ai missing key
* chore: fix sqlx errors in tests
* chore: document watcher - change plain text rendering to use depth-first scan
* chore: added nested document blocks example