feat(vectordb): 添加向量数据库支持及集成功能

新增向量数据库管理器模块，支持文本的存储、检索和相似度查询
添加知识库插件和AI聊天插件，利用向量数据库实现记忆功能
优化跨平台翻译模块，集成向量数据库存储历史翻译记录
改进消息处理逻辑，优先使用用户显示名称

plugins/ai_chat.py

@@ -0,0 +1,113 @@
+# -*- coding: utf-8 -*-
+"""
+AI 聊天插件，支持向量数据库记忆功能
+"""
+import time
+import uuid
+from core.managers.command_manager import matcher
+from models.events.message import GroupMessageEvent, PrivateMessageEvent
+from core.managers.vectordb_manager import vectordb_manager
+from core.utils.logger import ModuleLogger
+from core.config_loader import global_config
+
+logger = ModuleLogger("AIChat")
+
+# 尝试导入 OpenAI 客户端
+try:
+    from openai import AsyncOpenAI
+    OPENAI_AVAILABLE = True
+except ImportError:
+    OPENAI_AVAILABLE = False
+
+async def get_ai_response(user_id: int, group_id: int, user_message: str) -> str:
+    """获取 AI 回复，包含向量数据库记忆"""
+    if not OPENAI_AVAILABLE:
+        return "请先安装 openai 库: pip install openai"
+        
+    # 从配置中获取 DeepSeek API 配置（复用跨平台插件的配置或全局配置）
+    api_key = getattr(global_config.cross_platform, 'deepseek_api_key', None) or "your-api-key"
+    api_url = getattr(global_config.cross_platform, 'deepseek_api_url', "https://api.deepseek.com/v1")
+    model = getattr(global_config.cross_platform, 'deepseek_model', "deepseek-chat")
+    
+    if api_key == "your-api-key":
+        return "请先在配置中设置 DeepSeek API Key"
+        
+    # 1. 从向量数据库检索相关记忆
+    collection_name = f"chat_memory_{user_id}"
+    memory_context = ""
+    
+    try:
+        results = vectordb_manager.query_texts(
+            collection_name=collection_name,
+            query_texts=[user_message],
+            n_results=3
+        )
+        
+        if results and results.get("documents") and results["documents"][0]:
+            memory_context = "\n\n相关历史记忆：\n"
+            for i, doc in enumerate(results["documents"][0], 1):
+                memory_context += f"{i}. {doc}\n"
+    except Exception as e:
+        logger.error(f"检索聊天记忆失败: {e}")
+        
+    # 2. 构建 Prompt
+    system_prompt = f"""你是一个友好的 AI 助手。请根据用户的输入进行回复。
+如果提供了相关历史记忆，请参考这些记忆来保持对话的连贯性。{memory_context}"""
+
+    try:
+        client = AsyncOpenAI(
+            api_key=api_key,
+            base_url=api_url.replace("/chat/completions", "")
+        )
+        
+        response = await client.chat.completions.create(
+            model=model,
+            messages=[
+                {"role": "system", "content": system_prompt},
+                {"role": "user", "content": user_message}
+            ],
+            temperature=0.7,
+            max_tokens=1000
+        )
+        
+        ai_reply = response.choices[0].message.content
+        
+        # 3. 将本次对话存入向量数据库
+        if ai_reply:
+            try:
+                doc_id = str(uuid.uuid4())
+                text_to_embed = f"用户: {user_message}\nAI: {ai_reply}"
+                metadata = {
+                    "user_id": user_id,
+                    "group_id": group_id,
+                    "timestamp": int(time.time())
+                }
+                
+                vectordb_manager.add_texts(
+                    collection_name=collection_name,
+                    texts=[text_to_embed],
+                    metadatas=[metadata],
+                    ids=[doc_id]
+                )
+            except Exception as e:
+                logger.error(f"保存聊天记忆失败: {e}")
+                
+        return ai_reply
+    except Exception as e:
+        logger.error(f"AI 聊天请求失败: {e}")
+        return f"请求失败: {str(e)}"
+
+@matcher.command("chat", "聊天")
+async def chat_command(event: GroupMessageEvent | PrivateMessageEvent, args: list[str]):
+    """AI 聊天命令"""
+    if not args:
+        await event.reply("请提供要聊天的内容，例如：/chat 你好")
+        return
+        
+    user_message = " ".join(args)
+    user_id = event.user_id
+    group_id = getattr(event, 'group_id', 0)
+    
+    await event.reply("正在思考中...")
+    reply = await get_ai_response(user_id, group_id, user_message)
+    await event.reply(reply)