feat: add extra_body config for model-specific API parameters

Allows passing arbitrary parameters (e.g., enable_thinking, reasoning_effort)
to the LLM API request body via config.yaml, solving reasoning-only response
loops with models like Qwen 3.x without requiring code changes per model.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

config/config.yaml

@@ -10,6 +10,13 @@ llm:
   max_retries: 3
   retry_backoff_base: 1.0
   retry_backoff_max: 30.0
+  # Extra parameters merged into the API request body (model-specific).
+  # Examples:
+  #   Qwen 3.x:  enable_thinking: false
+  #   DeepSeek:   enable_thinking: false
+  #   OpenAI:     reasoning_effort: "low"
+  extra_body:
+    enable_thinking: false
 
 agent:
   max_iterations: 25