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/lib.rs

@@ -135,6 +135,7 @@ pub mod test_utils {
             client_manager,
             request_logger: Arc::new(crate::logging::RequestLogger::new(pool.clone(), dashboard_tx.clone())),
             model_registry: Arc::new(model_registry),
+            model_config_cache: crate::state::ModelConfigCache::new(pool.clone()),
             dashboard_tx,
             auth_tokens: vec![],
         })