feat: integrate vectordb for note embeddings

- Added `vectordb` as a dependency in `package.json`.
- Implemented `embedText` function in `src/main/ai/embeddings.ts` to handle text embeddings using GitHub Copilot OAuth token or OpenAI token.
- Created `vectordb.ts` for managing LanceDB connection and embedding notes with upsert strategy.
- Updated `index.ts` to initialize vector database and migrate existing notes on app ready.
- Modified `router/index.ts` to fire-and-forget embedding calls on note creation and updates.
- Enhanced `progress.txt` with detailed implementation notes and learnings regarding the integration.

src/main/router/index.ts

@@ -7,6 +7,7 @@ import { clients, projects, tasks, checkpoints, notes, taskComments } from '../d
 import { getStore } from '../store';
 import { saveTokenAndInit, hasActiveToken } from '../ai/provider';
 import { orchestrate } from '../ai/orchestrator';
+import { upsertNoteEmbedding } from '../db/vectordb';
 import type { TRPCContext } from '../ipc';
 
 const t = initTRPC.context<TRPCContext>().create();
@@ -406,7 +407,7 @@ const notesRouter = router({
 
   create: publicProcedure
     .input(z.object({ title: z.string(), content: z.string(), projectId: z.string().optional() }))
-    .mutation(({ input }) => {
+    .mutation(async ({ input }) => {
       const id = crypto.randomUUID();
       const now = Date.now();
       getDb().insert(notes).values({
@@ -417,18 +418,37 @@ const notesRouter = router({
         createdAt: now,
         updatedAt: now,
       }).run();
+      // Fire-and-forget: embed the note. Errors are logged, never thrown.
+      upsertNoteEmbedding(id, input.projectId ?? null, `${input.title}\n\n${input.content}`)
+        .catch((err) => console.error('[VectorDB] Failed to embed note on create:', err));
       return { id };
     }),
 
   update: publicProcedure
     .input(z.object({ id: z.string(), title: z.string().optional(), content: z.string().optional() }))
-    .mutation(({ input }) => {
+    .mutation(async ({ input }) => {
       const set: Partial<{ title: string; content: string; updatedAt: number }> = {};
       if (input.title !== undefined) set.title = input.title;
       if (input.content !== undefined) set.content = input.content;
       // Always update updatedAt
       set.updatedAt = Date.now();
       getDb().update(notes).set(set).where(eq(notes.id, input.id)).run();
+
+      // Re-embed if searchable text fields changed.
+      // Re-fetch from SQLite so the embedding reflects the full current note
+      // (the update may have changed only one of title or content).
+      if (input.title !== undefined || input.content !== undefined) {
+        const updated = getDb()
+          .select({ id: notes.id, projectId: notes.projectId, title: notes.title, content: notes.content })
+          .from(notes)
+          .where(eq(notes.id, input.id))
+          .all()[0];
+        if (updated) {
+          upsertNoteEmbedding(updated.id, updated.projectId ?? null, `${updated.title}\n\n${updated.content}`)
+            .catch((err) => console.error('[VectorDB] Failed to embed note on update:', err));
+        }
+      }
+
       return null;
     }),