feat: add Fireworks AI provider for chat agent

Routes chat-completions through a managed OpenAI-compatible inference
endpoint as an alternative to local Ollama, useful when the agent needs
fast multi-hop reasoning that local hardware can't sustain.

- backend: rename `call_ollama_chat_messages` → `call_chat_messages`,
  dispatch by provider; add `call_fireworks` branch (Bearer auth,
  `response_format: json_object` mapped from internal `format="json"`)
  and `list_fireworks_models` Tauri command
- settings: extend `AiProvider` enum + `AiSettings.fireworks_api_key`
  (serde-default for legacy config compat); Fireworks base URL hardcoded
- UI: provider selector in both popover and AppSettingsSheet (only
  ollama+fireworks shown; legacy openai/anthropic kept for serde-compat
  but normalized to ollama in UI); password input + dynamic model list
  for Fireworks; switching provider clears stale model selection
- 4 unit tests: serde round-trip, legacy settings deserialization,
  Fireworks chat-completions parsing, models-list parsing

src/lib/tauri.ts

@@ -211,6 +211,9 @@ export const saveAiSettings = (settings: AiSettings) =>
 export const listOllamaModels = (ollamaUrl: string) =>
   invoke<OllamaModel[]>("list_ollama_models", { ollamaUrl });
 
+export const listFireworksModels = (apiKey: string) =>
+  invoke<OllamaModel[]>("list_fireworks_models", { apiKey });
+
 export const generateSql = (connectionId: string, prompt: string) =>
   invoke<string>("generate_sql", { connectionId, prompt });