feat(llm): support configurations with YAML

[try-agaaain]

Link to #234

I'm working on replacing the .env configuration method with YAML and have completed a preliminary implementation. (I'll switch to OmegaConf later)

The generated configuration file looks like this:

config.yaml

AdminConfig:
  ADMIN_TOKEN: xxxx
  ENABLE_LOGIN: 'False'
  USER_TOKEN: '4321'
HugeGraphConfig:
  EDGE_LIMIT_PRE_LABEL: 8
  GRAPH_NAME: hugegraph
  GRAPH_PWD: xxx
  GRAPH_SPACE: null
  GRAPH_URL: 127.0.0.1:8080
  GRAPH_USER: admin
  LIMIT_PROPERTY: 'False'
  MAX_GRAPH_ITEMS: 30
  MAX_GRAPH_PATH: 10
  TOPK_PER_KEYWORD: 1
  TOPK_RETURN_RESULTS: 20
  VECTOR_DIS_THRESHOLD: 0.9
LLMConfig:
  CHAT_LLM_TYPE: openai
  COHERE_BASE_URL: https://api.cohere.com/v1/rerank
  EMBEDDING_TYPE: openai
  EXTRACT_LLM_TYPE: openai
  LITELLM_CHAT_API_BASE: null
  LITELLM_CHAT_API_KEY: null
  LITELLM_CHAT_LANGUAGE_MODEL: gemini-2.0-flash
  LITELLM_CHAT_TOKENS: 8192
  LITELLM_EMBEDDING_API_BASE: null
  LITELLM_EMBEDDING_API_KEY: null
  LITELLM_EMBEDDING_MODEL: openai/text-embedding-3-small
  LITELLM_EXTRACT_API_BASE: null
  LITELLM_EXTRACT_API_KEY: null
  LITELLM_EXTRACT_LANGUAGE_MODEL: gemini-2.0-flash
  LITELLM_EXTRACT_TOKENS: 256
  LITELLM_TEXT2GQL_API_BASE: null
  LITELLM_TEXT2GQL_API_KEY: null
  LITELLM_TEXT2GQL_LANGUAGE_MODEL: gemini-2.0-flash
  LITELLM_TEXT2GQL_TOKENS: 4096
  OLLAMA_CHAT_HOST: 127.0.0.1
  OLLAMA_CHAT_LANGUAGE_MODEL: null
  OLLAMA_CHAT_PORT: 11434
  OLLAMA_EMBEDDING_HOST: 127.0.0.1
  OLLAMA_EMBEDDING_MODEL: null
  OLLAMA_EMBEDDING_PORT: 11434
  OLLAMA_EXTRACT_HOST: 127.0.0.1
  OLLAMA_EXTRACT_LANGUAGE_MODEL: null
  OLLAMA_EXTRACT_PORT: 11434
  OLLAMA_TEXT2GQL_HOST: 127.0.0.1
  OLLAMA_TEXT2GQL_LANGUAGE_MODEL: null
  OLLAMA_TEXT2GQL_PORT: 11434
  OPENAI_CHAT_API_BASE: https://generativelanguage.googleapis.com/v1beta/openai
  OPENAI_CHAT_API_KEY: null
  OPENAI_CHAT_LANGUAGE_MODEL: gemini-2.0-flash
  OPENAI_CHAT_TOKENS: 8192
  OPENAI_EMBEDDING_API_BASE: https://generativelanguage.googleapis.com/v1beta/openai
  OPENAI_EMBEDDING_API_KEY: null
  OPENAI_EMBEDDING_MODEL: text-embedding-004
  OPENAI_EXTRACT_API_BASE: https://generativelanguage.googleapis.com/v1beta/openai
  OPENAI_EXTRACT_API_KEY: null
  OPENAI_EXTRACT_LANGUAGE_MODEL: gemini-2.0-flash
  OPENAI_EXTRACT_TOKENS: 8192
  OPENAI_TEXT2GQL_API_BASE: https://generativelanguage.googleapis.com/v1beta/openai
  OPENAI_TEXT2GQL_API_KEY: null
  OPENAI_TEXT2GQL_LANGUAGE_MODEL: gemini-2.0-flash
  OPENAI_TEXT2GQL_TOKENS: 8192
  QIANFAN_CHAT_ACCESS_TOKEN: null
  QIANFAN_CHAT_API_KEY: null
  QIANFAN_CHAT_LANGUAGE_MODEL: ERNIE-Speed-128K
  QIANFAN_CHAT_SECRET_KEY: null
  QIANFAN_CHAT_URL: https://aip.baidubce.com/rpc/2.0/ai_custom/v1/wenxinworkshop/chat/
  QIANFAN_EMBEDDING_API_KEY: null
  QIANFAN_EMBEDDING_MODEL: embedding-v1
  QIANFAN_EMBEDDING_SECRET_KEY: null
  QIANFAN_EMBED_URL: https://aip.baidubce.com/rpc/2.0/ai_custom/v1/wenxinworkshop/embeddings/
  QIANFAN_EXTRACT_ACCESS_TOKEN: null
  QIANFAN_EXTRACT_API_KEY: null
  QIANFAN_EXTRACT_LANGUAGE_MODEL: ERNIE-Speed-128K
  QIANFAN_EXTRACT_SECRET_KEY: null
  QIANFAN_TEXT2GQL_ACCESS_TOKEN: null
  QIANFAN_TEXT2GQL_API_KEY: null
  QIANFAN_TEXT2GQL_LANGUAGE_MODEL: ERNIE-Speed-128K
  QIANFAN_TEXT2GQL_SECRET_KEY: null
  QIANFAN_URL_PREFIX: https://aip.baidubce.com/rpc/2.0/ai_custom/v1/wenxinworkshop
  RERANKER_API_KEY: null
  RERANKER_MODEL: null
  RERANKER_TYPE: null
  TEXT2GQL_LLM_TYPE: openai

The LLMConfig section has a large number of configuration items. Do there any suggestions on how to better organize and manage these configurations?

[Copilot]

Pull Request Overview

This PR replaces the legacy .env configuration management with YAML-based configuration to streamline and centralize configuration handling. Key changes include the replacement of update_env calls with update_configs, updates to the BaseConfig class to support YAML generation and synchronization, and dependency updates across requirements and build files.

• Replaced .env-related functions with YAML-based functions in configuration blocks.
• Refactored BaseConfig to generate, update, and check YAML configurations.
• Updated project dependencies and .gitignore to support YAML.

Reviewed Changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
hugegraph-llm/src/hugegraph_llm/demo/rag_demo/configs_block.py	Replaced update_env with update_configs in configuration functions.
hugegraph-llm/src/hugegraph_llm/config/models/base_config.py	Refactored configuration syncing logic to use YAML instead of .env.
hugegraph-llm/requirements.txt	Added OmegaConf dependency.
hugegraph-llm/pyproject.toml	Updated dependencies (Gradio and OmegaConf).
hugegraph-llm/.gitignore	Configured to ignore config.yaml.

[Copilot]

Before iterating over the YAML config section, ensure that current_class_name exists as a key in yaml_config. Consider initializing yaml_config[current_class_name] to an empty dict if it does not exist to prevent potential KeyError.

Suggested change

	yaml_config = yaml.safe_load(content) if content.strip() else {}
	yaml_config = yaml.safe_load(content) if content.strip() else {}
	current_class_name = self.__class__.__name__
	if current_class_name not in yaml_config:
	yaml_config[current_class_name] = {}

[Copilot]

There is a minor typo in the comment ('onfig' should be 'config') – please correct it for clarity.

Suggested change

	# Step 2: Add missing onfig items from object to yaml.config
	# Step 2: Add missing config items from object to yaml.config

[imbajin]

Suggested change

	OmegaConf~=2.3
	OmegaConf~=2.3

keep a line -> EOF