fix: provider-aware model routing with fallback visibility

- Extract createClientFromConfig() to dispatch on provider field instead
  of hardcoding all tiers as AnthropicClient
- Add fallback/fallbackReason metadata to ChatResponse and ChatStreamEvent
  so callers know when a fallback model was used
- Enhance doctor check to report full model stack and warn on missing
  API keys for cloud providers
- Log fallback warnings in NativeAgent and display them in TUI
- Support tier names and local_providers entries in fallback_chain
- Add 8 tests for createClientFromConfig covering all provider types

src/models/router.test.ts

@@ -41,6 +41,8 @@ describe('ModelRouter', () => {
     const response = await router.chat({ messages: [{ role: 'user', content: 'Hi' }] });
 
     expect(response.content).toBe('Response from fallback');
+    expect(response.fallback).toBe(true);
+    expect(response.fallbackReason).toMatch(/Primary model failed/);
     expect(failingClient.chat).toHaveBeenCalled();
     expect(fallbackClient.chat).toHaveBeenCalled();
   });
@@ -132,13 +134,18 @@ describe('ModelRouter streaming', () => {
     });
 
     const chunks: string[] = [];
+    let fallbackWarning: string | undefined;
     for await (const event of router.chatStream({ messages: [] })) {
       if (event.type === 'content' && event.content) {
         chunks.push(event.content);
       }
+      if (event.type === 'fallback_warning') {
+        fallbackWarning = event.fallbackReason;
+      }
     }
 
     expect(chunks).toEqual(['Fallback']);
+    expect(fallbackWarning).toMatch(/Primary model failed/);
   });
 });