feat: Migrate document parsing to Aliyun and update embedding configurations

- Updated LocalDocumentParser to include raw_layouts and artifact_prefix from settings.
- Added new documents with failure reasons and metadata to documents.json for better error tracking.
- Created a new documentation file detailing the Aliyun ingest implementation process.
- Updated RFC to reflect changes in the parsing backend and embedding dimensions.
- Modified tests to accommodate the new embedding dimension of 1024 and updated parser and chunk builder assertions.
- Verified migration configurations to ensure correct settings for embedding model and backend.

backend/app/config/settings.py

@@ -1,9 +1,9 @@
-"""Configure backend settings for settings."""
+"""Configure backend settings for the backend application."""
 
 from pathlib import Path
 
-from pydantic_settings import BaseSettings, SettingsConfigDict
 from pydantic import Field
+from pydantic_settings import BaseSettings, SettingsConfigDict
 from functools import lru_cache
 # Keep configuration setup explicit so runtime behavior is easy to reason about.
 
@@ -33,18 +33,25 @@ class Settings(BaseSettings):
     # Keep configuration setup explicit so runtime behavior is easy to reason about.
     milvus_host: str = Field(default="localhost", description="Milvus服务地址")
     milvus_port: int = Field(default=19530, description="Milvus服务端口")
-    milvus_collection: str = Field(default="regulations_dense_1536", description="法规向量集合名称")
+    milvus_collection: str = Field(default="regulations_dense_1024_v1", description="法规向量集合名称")
     milvus_db_name: str = Field(default="default", description="Milvus数据库名称")
 
     # Keep configuration setup explicit so runtime behavior is easy to reason about.
     embedding_model: str = Field(default="text-embedding-v3", description="嵌入模型名称")
-    embedding_dim: int = Field(default=1536, description="嵌入向量维度")
+    embedding_dim: int = Field(default=1024, description="嵌入向量维度")
     embedding_api_key: str = Field(default="", description="Embedding API密钥")
     embedding_base_url: str = Field(default="http://6.86.80.4:30080/v1", description="Embedding API地址")
     embedding_timeout_seconds: int = Field(default=120, description="Embedding API超时时间(秒)")
     alibaba_access_key_id: str = Field(default="", description="阿里云文档解析 Access Key ID")
     alibaba_access_key_secret: str = Field(default="", description="阿里云文档解析 Access Key Secret")
     alibaba_endpoint: str = Field(default="docmind-api.cn-hangzhou.aliyuncs.com", description="阿里云文档解析 endpoint")
+    aliyun_parse_poll_interval_seconds: int = Field(default=5, description="阿里云文档解析轮询间隔(秒)")
+    aliyun_parse_timeout_seconds: int = Field(default=900, description="阿里云文档解析超时时间(秒)")
+    aliyun_parse_layout_step_size: int = Field(default=50, description="阿里云文档解析分页步长")
+    aliyun_llm_enhancement: bool = Field(default=True, description="是否启用阿里云解析增强")
+    aliyun_enhancement_mode: str = Field(default="VLM", description="阿里云解析增强模式")
+    document_parse_artifact_prefix: str = Field(default="artifacts", description="解析产物对象前缀")
+    parser_failure_mode: str = Field(default="fail", description="解析失败策略")
 
     # Keep configuration setup explicit so runtime behavior is easy to reason about.
     minio_endpoint: str = Field(default="localhost:9000", description="MinIO服务地址")
@@ -71,8 +78,8 @@ class Settings(BaseSettings):
     chunk_overlap: int = Field(default=50, description="分块重叠大小")
     max_file_size_mb: int = Field(default=100, description="最大文件大小(MB)")
     document_metadata_path: str = Field(default="backend/data/documents.json", description="文档元数据存储路径")
-    parser_backend: str = Field(default="local", description="解析后端(local/aliyun)")
-    chunk_backend: str = Field(default="local", description="分块后端(local/aliyun)")
+    parser_backend: str = Field(default="aliyun", description="解析后端(local/aliyun)")
+    chunk_backend: str = Field(default="aliyun", description="分块后端(local/aliyun)")
 
     # Keep configuration setup explicit so runtime behavior is easy to reason about.
     api_host: str = Field(default="0.0.0.0", description="API服务地址")