Add fallback models when not configured

When no models are specified in agent config, use sensible defaults:
- Chat/Default tasks: ollama/llama3
- Code/Planning/ToolUse: openai/gpt-4o

This allows minimal configs without explicit model routing.

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

src/agents/agent.rs

@@ -76,9 +76,23 @@ impl Agent {
 
     /// Get the model for a specific task type
     fn get_model(&self, task_type: TaskType) -> Result<String> {
-        self.config
-            .get_model_for_task(task_type)
-            .ok_or_else(|| AgentError::ConfigError(format!("No model configured for {:?}", task_type)))
+        // First try to get model from agent config
+        if let Some(model) = self.config.get_model_for_task(task_type) {
+            return Ok(model);
+        }
+        
+        // Fallback: use a reasonable default based on available providers
+        // Prefer ollama for local-first, then openai
+        match task_type {
+            TaskType::Code | TaskType::Planning | TaskType::ToolUse => {
+                // For complex tasks, try openai first
+                Ok("openai/gpt-4o".to_string())
+            }
+            TaskType::Chat | TaskType::Default => {
+                // For chat and default, use ollama
+                Ok("ollama/llama3".to_string())
+            }
+        }
     }
 
     /// Get available tools as provider Tool definitions