feat: add AnthropicVertexClient for claude-* models

- @anthropic-ai/vertex-sdk: proper Anthropic Messages API on Vertex
- AnthropicVertexClient: converts OAI message format ↔ Anthropic format,
  handles tool_use blocks, retries 429/503 with backoff
- createLLM: routes anthropic/* and claude-* models through new client
- Tier B/C default: claude-sonnet-4-6 via us-east5 Vertex endpoint
- /generate endpoint: accepts region param for regional endpoint testing

Made-with: Cursor

src/server.ts

@@ -550,18 +550,24 @@ app.post('/agent/approve', async (req: Request, res: Response) => {
 // ---------------------------------------------------------------------------
 
 app.post('/generate', async (req: Request, res: Response) => {
-    const { prompt, model } = req.body as { prompt?: string; model?: string };
+    const { prompt, model, region } = req.body as { prompt?: string; model?: string; region?: string };
     if (!prompt) { res.status(400).json({ error: '"prompt" is required' }); return; }
 
+    // Allow overriding CLAUDE_REGION per-request for testing
+    const prevRegion = process.env.CLAUDE_REGION;
+    if (region) process.env.CLAUDE_REGION = region;
+
     try {
         const llm = createLLM(model ?? 'A', { temperature: 0.3 });
         const messages: import('./llm').LLMMessage[] = [
             { role: 'user', content: prompt }
         ];
         const response = await llm.chat(messages, [], 8192);
-        res.json({ reply: response.content ?? '' });
+        res.json({ reply: response.content ?? '', model: llm.modelId });
     } catch (err) {
-        res.status(500).json({ error: err instanceof Error ? err.message : String(err) });
+        res.status(500).json({ error: err instanceof Error ? err.message : String(err), model });
+    } finally {
+        if (region) process.env.CLAUDE_REGION = prevRegion ?? '';
     }
 });