perf: eliminate per-request SQLite queries and optimize proxy latency

- Add in-memory ModelConfigCache (30s refresh, explicit invalidation)
  replacing 2 SQLite queries per request (model lookup + cost override)
- Configure all 5 provider HTTP clients with proper timeouts (300s),
  connection pooling (4 idle/host, 90s idle timeout), and TCP keepalive
- Move client_usage update to tokio::spawn in non-streaming path
- Use fast chars/4 heuristic for token estimation on large inputs (>1KB)
- Generate single UUID/timestamp per SSE stream instead of per chunk
- Add shared LazyLock<Client> for image fetching in multimodal module
- Add proxy overhead timing instrumentation for both request paths
- Fix test helper to include new model_config_cache field

src/main.rs

@@ -65,6 +65,11 @@ async fn main() -> Result<()> {
         config.server.auth_tokens.clone(),
     );
 
+    // Initialize model config cache and start background refresh (every 30s)
+    state.model_config_cache.refresh().await;
+    state.model_config_cache.clone().start_refresh_task(30);
+    info!("Model config cache initialized");
+
     // Create application router
     let app = Router::new()
         .route("/health", get(health_check))