feat(planner): 实现大小脑规划器分离以优化决策流程

将规划器（Planner）拆分为“大脑”和“小脑”两个部分，以实现更精细化的决策控制。

- **大脑（BIG_BRAIN）**: 负责宏观决策，如是否回复、是否需要@人等高层级意图。
- **小脑（SMALL_BRAIN）**: 负责具体的功能性动作执行。

此重构引入了 `PlannerType` 枚举，并更新了动作（Action）定义，允许将动作明确分配给大脑或小脑，从而提升了AI回复的逻辑性和条理性。同时，新增了 `no_action` 类型，用于在规划阶段明确表示“无动作”，提高了处理流程的清晰度。

src/chat/chat_loop/cycle_processor.py

@@ -228,6 +228,8 @@ class CycleProcessor:
         async def execute_action(action_info):
             """执行单个动作的通用函数"""
             try:
+                if action_info["action_type"] == "no_action":
+                    return {"action_type": "no_action", "success": True, "reply_text": "", "command": ""}            
                 if action_info["action_type"] == "no_reply":
                     # 直接处理no_reply逻辑，不再通过动作系统
                     reason = action_info.get("reasoning", "选择不回复")
@@ -245,7 +247,7 @@ class CycleProcessor:
                     )
 
                     return {"action_type": "no_reply", "success": True, "reply_text": "", "command": ""}
-                elif action_info["action_type"] != "reply":
+                elif action_info["action_type"] != "reply" and action_info["action_type"] != "no_action":
                     # 执行普通动作
                     with Timer("动作执行", cycle_timers):
                         success, reply_text, command = await self._handle_action(
@@ -420,7 +422,7 @@ class CycleProcessor:
                 if "reply" in available_actions:
                     fallback_action = "reply"
                 elif available_actions:
-                    fallback_action = list(available_actions.keys())
+                    fallback_action = list(available_actions.keys())[0]
 
                 if fallback_action and fallback_action != action:
                     logger.info(f"{self.context.log_prefix} 使用回退动作: {fallback_action}")

src/chat/planner_actions/planner.py

@@ -23,7 +23,13 @@ from src.chat.utils.chat_message_builder import (
 from src.chat.utils.utils import get_chat_type_and_target_info
 from src.chat.planner_actions.action_manager import ActionManager
 from src.chat.message_receive.chat_stream import get_chat_manager
-from src.plugin_system.base.component_types import ActionInfo, ChatMode, ComponentType, ActionActivationType
+from src.plugin_system.base.component_types import (
+    ActionInfo,
+    ChatMode,
+    ComponentType,
+    ActionActivationType,
+    PlannerType,
+)
 from src.plugin_system.core.component_registry import component_registry
 from src.schedule.schedule_manager import schedule_manager
 from src.mood.mood_manager import mood_manager
@@ -503,6 +509,9 @@ class ActionPlanner:
         try:
             sub_planner_actions: Dict[str, ActionInfo] = {}
             for action_name, action_info in available_actions.items():
+                if action_info.planner_type not in [PlannerType.SMALL_BRAIN, PlannerType.ALL]:
+                    continue
+
                 if action_info.activation_type in [ActionActivationType.LLM_JUDGE, ActionActivationType.ALWAYS]:
                     sub_planner_actions[action_name] = action_info
                 elif action_info.activation_type == ActionActivationType.RANDOM:
@@ -550,10 +559,15 @@ class ActionPlanner:
         # --- 3. 大脑独立思考是否回复 ---
         action, reasoning, action_data, target_message = "no_reply", "大脑初始化默认", {}, None
         try:
+            big_brain_actions = {
+                name: info
+                for name, info in available_actions.items()
+                if info.planner_type in [PlannerType.BIG_BRAIN, PlannerType.ALL]
+            }
             prompt, _ = await self.build_planner_prompt(
                 is_group_chat=is_group_chat,
                 chat_target_info=chat_target_info,
-                current_available_actions={},  # 大脑不考虑具体action
+                current_available_actions=big_brain_actions,
                 mode=mode,
                 chat_content_block_override=chat_content_block,
                 message_id_list_override=message_id_list,