feat: Step 6 - Write-through cache API

Implement the core write-through cache pattern for node creation.
This is the architectural foundation of the application.

Changes:
- Add @google/generative-ai dependency for embeddings
- Create lib/db.ts: SurrealDB connection helper with JWT auth
- Create lib/ai.ts: AI embedding generation using text-embedding-004
- Create app/api/nodes/route.ts: POST endpoint implementing write-through cache

Write-through cache flow:
1. Authenticate user via SurrealDB JWT
2. Publish node to ATproto PDS (source of truth)
3. Generate 768-dimensional embedding via Google AI
4. Cache node + embedding + links in SurrealDB

Updated schema to use 768-dimensional embeddings (text-embedding-004)
instead of 1536 dimensions.

Security:
- Row-level permissions enforced via SurrealDB JWT
- All secrets server-side only
- ATproto OAuth tokens from secure cookies

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

db/schema.surql

@@ -68,9 +68,9 @@ DEFINE FIELD coords_3d ON TABLE node TYPE array<number>
 
 -- Define the vector search index.
 -- We use MTREE (or HNSW) for high-performance k-NN search.
--- The dimension (1536) MUST match the output of the
--- 'gemini-embedding-001' model.
-DEFINE INDEX node_embedding_idx ON TABLE node FIELDS embedding MTREE DIMENSION 1536;
+-- The dimension (768) MUST match the output of the
+-- 'text-embedding-004' model.
+DEFINE INDEX node_embedding_idx ON TABLE node FIELDS embedding MTREE DIMENSION 768;
 
 -- --------------------------------------------------
 -- Relation: links_to