feat(arr): make the OpenAI endpoint a live setting

The base URL and model move into a cell the backend re-reads per request,
so an operator can repoint it without a restart. Migration 0028 adds the
two columns; DESIGN.md §15 calls both database rows. Construction never
depends on the API key — llama.cpp serves without one.

crates/arr-db/migrations/0028_openai_endpoint.sql

@@ -0,0 +1,17 @@
+-- #220. Where the OpenAI-compatible translation backend points, and which
+-- model it names (DESIGN.md §15). That backend is not "OpenAI" — it is
+-- anything speaking that request shape, `llama.cpp` and a local gateway
+-- included — so which endpoint and which model are in use is something the
+-- operator tries and changes, not a property of the deployment fixed at
+-- start-up. Both replace bootstrap keys #216 added as an explicit stopgap.
+--
+-- Only the API key stays in the environment (§10): it is a secret, and an
+-- endpoint that needs no key at all is a valid configuration.
+
+ALTER TABLE subtitle_settings
+    -- NULL means the backend's own default, `https://api.openai.com/v1/`.
+    ADD COLUMN openai_base_url TEXT;
+
+ALTER TABLE subtitle_settings
+    -- NULL means the backend's own default model.
+    ADD COLUMN openai_model TEXT;