feat(annotation): 自动标注任务支持非图像类型数据集（TEXT/AUDIO/VIDEO）

移除自动标注任务创建流程中的 IMAGE-only 限制，使 TEXT、AUDIO、VIDEO
类型数据集均可用于自动标注任务。

- 新增数据库迁移：t_dm_auto_annotation_tasks 表添加 dataset_type 列
- 后端 schema/API/service 全链路传递 dataset_type
- Worker 动态构建 sample key（image/text/audio/video）和输出目录
- 前端移除数据集类型校验，下拉框显示数据集类型标识
- 输出数据集继承源数据集类型，不再硬编码为 IMAGE
- 保持向后兼容：默认值为 IMAGE，worker 有元数据回退和目录 fallback

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

runtime/datamate-python/app/module/annotation/schema/auto.py

@@ -1,4 +1,4 @@
-"""Schemas for Auto Annotation tasks"""
+"""Schemas for Auto Annotation tasks"""
 from __future__ import annotations
 
 import json
@@ -7,24 +7,24 @@ from typing import List, Optional, Dict, Any
 from datetime import datetime
 
 from pydantic import BaseModel, Field, ConfigDict, model_validator
-
-
+
+
 class AutoAnnotationConfig(BaseModel):
-    """自动标注任务配置（与前端 payload 对齐）"""
-
-    model_size: str = Field(alias="modelSize", description="模型规模: n/s/m/l/x")
-    conf_threshold: float = Field(alias="confThreshold", description="置信度阈值 0-1")
-    target_classes: List[int] = Field(
-        default_factory=list,
-        alias="targetClasses",
-        description="目标类别ID列表，空表示全部类别",
-    )
-    output_dataset_name: Optional[str] = Field(
-        default=None,
-        alias="outputDatasetName",
-        description="自动标注结果要写入的新数据集名称（可选）",
-    )
-
+    """自动标注任务配置（与前端 payload 对齐）"""
+
+    model_size: str = Field(alias="modelSize", description="模型规模: n/s/m/l/x")
+    conf_threshold: float = Field(alias="confThreshold", description="置信度阈值 0-1")
+    target_classes: List[int] = Field(
+        default_factory=list,
+        alias="targetClasses",
+        description="目标类别ID列表，空表示全部类别",
+    )
+    output_dataset_name: Optional[str] = Field(
+        default=None,
+        alias="outputDatasetName",
+        description="自动标注结果要写入的新数据集名称（可选）",
+    )
+
     model_config = ConfigDict(populate_by_name=True)
 
 
@@ -68,13 +68,18 @@ class OperatorPipelineStep(BaseModel):
         return normalized
 
     model_config = ConfigDict(populate_by_name=True)
-
-
+
+
 class CreateAutoAnnotationTaskRequest(BaseModel):
     """创建自动标注任务的请求体，对齐前端 CreateAutoAnnotationDialog 发送的结构"""
 
     name: str = Field(..., min_length=1, max_length=255, description="任务名称")
     dataset_id: str = Field(..., alias="datasetId", description="数据集ID")
+    dataset_type: Optional[str] = Field(
+        default=None,
+        alias="datasetType",
+        description="数据集类型: IMAGE/TEXT/AUDIO/VIDEO（不传时由后端自动获取）",
+    )
     config: Optional[AutoAnnotationConfig] = Field(
         default=None,
         description="兼容旧版 YOLO 任务配置",
@@ -111,15 +116,16 @@ class CreateAutoAnnotationTaskRequest(BaseModel):
         return self
 
     model_config = ConfigDict(populate_by_name=True)
-
-
+
+
 class AutoAnnotationTaskResponse(BaseModel):
-    """自动标注任务响应模型（列表/详情均可复用）"""
-
-    id: str = Field(..., description="任务ID")
-    name: str = Field(..., description="任务名称")
-    dataset_id: str = Field(..., alias="datasetId", description="数据集ID")
+    """自动标注任务响应模型（列表/详情均可复用）"""
+
+    id: str = Field(..., description="任务ID")
+    name: str = Field(..., description="任务名称")
+    dataset_id: str = Field(..., alias="datasetId", description="数据集ID")
     dataset_name: Optional[str] = Field(None, alias="datasetName", description="数据集名称")
+    dataset_type: Optional[str] = Field(None, alias="datasetType", description="数据集类型: IMAGE/TEXT/AUDIO/VIDEO")
     task_mode: Optional[str] = Field(None, alias="taskMode", description="任务模式")
     executor_type: Optional[str] = Field(None, alias="executorType", description="执行器类型")
     pipeline: Optional[List[Dict[str, Any]]] = Field(None, description="算子编排定义")
@@ -128,11 +134,11 @@ class AutoAnnotationTaskResponse(BaseModel):
         alias="sourceDatasets",
         description="本任务实际处理涉及到的所有数据集名称列表",
     )
-    config: Dict[str, Any] = Field(..., description="任务配置")
-    status: str = Field(..., description="任务状态")
-    progress: int = Field(..., description="任务进度 0-100")
-    total_images: int = Field(..., alias="totalImages", description="总图片数")
-    processed_images: int = Field(..., alias="processedImages", description="已处理图片数")
+    config: Dict[str, Any] = Field(..., description="任务配置")
+    status: str = Field(..., description="任务状态")
+    progress: int = Field(..., description="任务进度 0-100")
+    total_images: int = Field(..., alias="totalImages", description="总图片数")
+    processed_images: int = Field(..., alias="processedImages", description="已处理图片数")
     detected_objects: int = Field(..., alias="detectedObjects", description="检测到的对象总数")
     output_path: Optional[str] = Field(None, alias="outputPath", description="输出路径")
     output_dataset_id: Optional[str] = Field(
@@ -152,14 +158,14 @@ class AutoAnnotationTaskResponse(BaseModel):
     created_at: datetime = Field(..., alias="createdAt", description="创建时间")
     updated_at: Optional[datetime] = Field(None, alias="updatedAt", description="更新时间")
     completed_at: Optional[datetime] = Field(None, alias="completedAt", description="完成时间")
-
-    model_config = ConfigDict(populate_by_name=True, from_attributes=True)
-
-
-class AutoAnnotationTaskListResponse(BaseModel):
-    """自动标注任务列表响应，目前前端直接使用数组，这里预留分页结构"""
-
-    content: List[AutoAnnotationTaskResponse] = Field(..., description="任务列表")
-    total: int = Field(..., description="总数")
-
-    model_config = ConfigDict(populate_by_name=True)
+
+    model_config = ConfigDict(populate_by_name=True, from_attributes=True)
+
+
+class AutoAnnotationTaskListResponse(BaseModel):
+    """自动标注任务列表响应，目前前端直接使用数组，这里预留分页结构"""
+
+    content: List[AutoAnnotationTaskResponse] = Field(..., description="任务列表")
+    total: int = Field(..., description="总数")
+
+    model_config = ConfigDict(populate_by_name=True)