修复AI建议逻辑和字段映射问题

- 修复AI建议基于问题描述而不是处理过程生成
- 修复工单详情页面显示逻辑
- 修复飞书时间字段处理（毫秒时间戳转换）
- 优化字段映射和转换逻辑
- 添加飞书集成功能
- 改进对话历史合并功能
- 优化系统优化反馈机制

src/dialogue/dialogue_manager.py

@@ -101,26 +101,47 @@ class DialogueManager:
             # 性能优化分析
             optimization_result = self.system_optimizer.optimize_response_time(response_time)
             
-            # 记录Token使用情况
-            if success and "token_usage" in response_result:
-                token_usage = response_result["token_usage"]
-                # 计算成本
+            # 记录Token使用情况（兼容多种返回格式）
+            if success:
+                # 兼容返回 usage: {prompt_tokens, completion_tokens}
+                usage = response_result.get("usage", {}) or {}
+                token_usage = response_result.get("token_usage", {}) or {}
+                input_tokens = token_usage.get("input_tokens")
+                output_tokens = token_usage.get("output_tokens")
+                if input_tokens is None and isinstance(usage, dict):
+                    input_tokens = usage.get("prompt_tokens") or usage.get("input_tokens") or 0
+                if output_tokens is None and isinstance(usage, dict):
+                    output_tokens = usage.get("completion_tokens") or usage.get("output_tokens") or 0
+
+                # 若均为0，使用简易估算（避免记录缺失）
+                if not input_tokens and user_message:
+                    try:
+                        input_tokens = max(1, len(user_message) // 4)
+                    except Exception:
+                        input_tokens = 0
+                if not output_tokens and response_result.get("response"):
+                    try:
+                        output_tokens = max(1, len(response_result.get("response")) // 4)
+                    except Exception:
+                        output_tokens = 0
+
+                model_name = response_result.get("model") or response_result.get("model_name") or "qwen-plus-latest"
+
+                # 计算成本并限制
                 estimated_cost = self.token_monitor._calculate_cost(
-                    response_result.get("model_name", "qwen-plus-latest"),
-                    token_usage.get("input_tokens", 0),
-                    token_usage.get("output_tokens", 0)
+                    model_name,
+                    int(input_tokens or 0),
+                    int(output_tokens or 0)
                 )
-                
-                # 检查成本限制
                 if not self.system_optimizer.check_cost_limit(estimated_cost):
                     return {"error": "请求成本超限，请稍后再试"}
-                
+
                 self.token_monitor.record_token_usage(
                     user_id=user_id or "anonymous",
                     work_order_id=work_order_id,
-                    model_name=response_result.get("model_name", "qwen-plus-latest"),
-                    input_tokens=token_usage.get("input_tokens", 0),
-                    output_tokens=token_usage.get("output_tokens", 0),
+                    model_name=model_name,
+                    input_tokens=int(input_tokens or 0),
+                    output_tokens=int(output_tokens or 0),
                     response_time=response_time,
                     success=success,
                     error_message=error_message