Route all AI features through the Switchboard gateway

Replace the direct Anthropic and OpenAI integrations with a single
provider that talks to Switchboard, an OpenAI-compatible gateway that
routes each request to the best available model. The app no longer pins
a model id anywhere: it sends switchboard/auto and lets the gateway
choose, then logs which model answered and what it cost.

Routing levers are set per feature in src/lib/ai/routing.ts. Three of
those choices came from measuring against the live gateway:

- category and prefer_free are set explicitly on every request. An API
  key carries its own routing defaults, and anything left unset inherits
  them - drink prompts were being sent to a free coding model.
- Token budgets are generous because the router may pick a reasoning
  model, and reasoning tokens come out of the same max_tokens budget as
  the answer. At 512 tokens a request returned null content; at 4096 the
  same request returned correct JSON.
- No tier lever on text features. tier "cheap" pinned a slow reasoning
  model (42-180s, two timeouts and one truncated response in five
  trials) and tier "frontier" escalated as far as Opus at $0.02 a call,
  while unconstrained routing answered in about a second. Vision keeps
  "frontier", where the accuracy is worth a few tenths of a cent.

Gateway failures are mapped to actionable messages rather than passed
through: a 401 relayed as 401 would read as an expired session and
bounce the user to login, and a 429 would collide with the app's own
rate limiter.

Also collapses the key lookup that was duplicated across ten call sites
into getUserProvider(), which fixes a latent bug where a bare findFirst
with no ordering let different features pick different providers.

Existing claude/openai key rows are ignored at runtime and offered for
removal in Settings, so no migration is needed before deploying.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

README.md

@@ -20,6 +20,37 @@ You can start editing the page by modifying `app/page.tsx`. The page auto-update
 
 This project uses [`next/font`](https://nextjs.org/docs/app/building-your-application/optimizing/fonts) to automatically optimize and load [Geist](https://vercel.com/font), a new font family for Vercel.
 
+## AI Gateway (Switchboard)
+
+All AI features — menu scanning, label identification, drink search, the bartender
+and the recommendation engine — go through [Switchboard](http://192.168.2.11:8787/v1/guide),
+an OpenAI-compatible gateway that routes each request to the best available model.
+The app never pins a model id; it always sends `switchboard/auto` and lets the gateway
+choose, then logs which model answered and what it cost.
+
+Setup:
+
+1. Set `SWITCHBOARD_BASE_URL` in your env file (defaults to `http://192.168.2.11:8787/v1`).
+2. Mint an API key in the Switchboard UI under Settings → API keys.
+3. Add that key in the app under Settings → AI Gateway.
+
+Per-feature routing (cost/quality levers, token budgets, timeouts) lives in
+`src/lib/ai/routing.ts`. Note that a Switchboard key carries its own routing defaults,
+so the app sets `category` and `prefer_free` explicitly on every request rather than
+inheriting whatever the key was minted for.
+
+### Migrating from the old Claude/OpenAI integration
+
+Earlier versions stored a per-user Anthropic or OpenAI key. Those rows are ignored at
+runtime and the Settings page offers to remove them, so no migration is required. To
+clear them in bulk instead:
+
+```sql
+DELETE FROM "UserApiKey"  WHERE provider IN ('claude','openai');
+DELETE FROM "SearchCache" WHERE provider IN ('claude','openai');
+UPDATE "UserPreference" SET "defaultProvider" = NULL;
+```
+
 ## Learn More
 
 To learn more about Next.js, take a look at the following resources: