feat: integrate Langfuse for LLM observability

- Add instrumentation.ts with Langfuse OpenTelemetry exporter
- Enable experimental telemetry on streamText calls
- Add instrumentationHook to Next.js config
- Install required dependencies (@vercel/otel, langfuse-vercel, etc.)

package.json

@@ -18,17 +18,22 @@
         "@ai-sdk/react": "^2.0.22",
         "@next/third-parties": "^16.0.6",
         "@openrouter/ai-sdk-provider": "^1.2.3",
+        "@opentelemetry/api-logs": "^0.208.0",
+        "@opentelemetry/instrumentation": "^0.208.0",
+        "@opentelemetry/sdk-logs": "^0.208.0",
         "@radix-ui/react-dialog": "^1.1.6",
         "@radix-ui/react-scroll-area": "^1.2.3",
         "@radix-ui/react-slot": "^1.1.2",
         "@radix-ui/react-tooltip": "^1.1.8",
         "@vercel/analytics": "^1.5.0",
+        "@vercel/otel": "^2.1.0",
         "@xmldom/xmldom": "^0.9.8",
         "ai": "^5.0.89",
         "base-64": "^1.0.0",
         "class-variance-authority": "^0.7.1",
         "clsx": "^2.1.1",
         "jsdom": "^26.0.0",
+        "langfuse-vercel": "^3.38.6",
         "lucide-react": "^0.483.0",
         "next": "15.2.3",
         "ollama-ai-provider-v2": "^1.5.4",