refactor(ui): use tokenlens as sole model source, remove provider listModels

Remove the per-provider listModels API (GET /providers/:id/models) and all
four provider implementations (OpenAI Compatible, OpenAI Responses, Anthropic,
Gemini). ModelCombobox now only shows tokenlens suggestions (tagged '推荐') plus
free-form custom input — no more unfiltered 'API' models from provider SDKs.

Fixes: switching provider type in ProviderDialog no longer shows stale models
from the original provider's API.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode)

src/controllers/llm-config.ts

@@ -55,29 +55,6 @@ llmConfigRouter.get('/providers/:id', (c) => {
   });
 });
 
-llmConfigRouter.get('/providers/:id/models', async (c) => {
-  const id = c.req.param('id');
-  const provider = providerRepo.getById(id);
-  if (!provider) return c.json({ message: 'Provider not found' }, 404);
-
-  if (!secretRepo.has(id)) {
-    return c.json({ message: 'No API key configured' }, 400);
-  }
-
-  try {
-    llmGateway.invalidateProvider(id);
-    const providerInstance = llmGateway.getProviderInstance(id);
-    if (!providerInstance.listModels) {
-      return c.json({ message: 'This provider does not support listing models' }, 501);
-    }
-
-    const models = await providerInstance.listModels();
-    return c.json({ models });
-  } catch (error: any) {
-    return c.json({ message: error.message || 'Failed to fetch models' }, 500);
-  }
-});
-
 llmConfigRouter.post('/providers', async (c) => {
   const body = await c.req.json<{
     name: string;