fix: add 5s connect_timeout to LLM HTTP client so unreachable APIs fail fast

Without connect_timeout, reqwest can hang for the full 60s timeout
when the remote server is unreachable (DNS, TCP connect, etc.).
Now fails in 5s max for connection issues, 30s for full request.

This means one user's LLM failure no longer blocks new users for
a full minute — the channel closes quickly and the WebSocket is freed.

Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>

src/llm/mod.rs

@@ -200,7 +200,8 @@ impl OpenAIClient {
 
         Self {
             client: reqwest::Client::builder()
-                .timeout(Duration::from_secs(60))
+                .connect_timeout(Duration::from_secs(5))
+                .timeout(Duration::from_secs(30))
                 .build()
                 .unwrap_or_else(|_| reqwest::Client::new()),
             base_url: base,