feat(workflow, skill): add multimodal image support to workflows and skill prompt generation

api/app/schemas/app_schema.py

@@ -1,14 +1,14 @@
 import datetime
 import uuid
 from typing import Optional, Any, List, Dict, Union
-from enum import Enum
+from enum import Enum, StrEnum
 
 from pydantic import BaseModel, Field, ConfigDict, field_serializer, field_validator
 
 
 # ---------- Multimodal File Support ----------
 
-class FileType(str, Enum):
+class FileType(StrEnum):
     """文件类型枚举"""
     IMAGE = "image"
     DOCUMENT = "document"

api/app/schemas/prompt_optimizer_schema.py

@@ -21,6 +21,11 @@ class PromptOptMessage(BaseModel):
         description="currently optimized prompt"
     )
 
+    skill: bool = Field(
+        default=False,
+        description="Enable variable output"
+    )
+
 
 class PromptSaveRequest(BaseModel):
     session_id: UUID = Field(