**feat:** update default model to qwen3:8b and simplify chat streaming loop with proper error handling and trailing newline.

crates/cli/src/main.rs

@@ -52,7 +52,7 @@ async fn main() -> Result<()> {
     }];
 
     let mut stream = client.chat_stream(&msgs, &opts).await?;
-    while let Ok(Some(chunk)) = stream.try_next().await {
+    while let Some(chunk) = stream.try_next().await? {
         if let Some(m) = chunk.message {
             if let Some(c) = m.content {
                 print!("{c}");
@@ -63,5 +63,6 @@ async fn main() -> Result<()> {
             break;
         }
     }
+    println!(); // Newline after response
     Ok(())
 }

crates/config/src/lib.rs

@@ -20,7 +20,7 @@ fn default_ollama_url() -> String {
     "http://localhost:11434".into()
 }
 fn default_model() -> String {
-    "qwen2.5".into()
+    "qwen3:8b".into()
 }
 fn default_mode() -> String {
     "plan".into()