feat: BaseTextQuality根据score来判断

[shijinpjlab]

No description provided.

[gemini-code-assist]

Code Review

This pull request updates the text quality evaluation logic in base_text_quality.py to determine status and labels based on a numerical score of 1 rather than a type string. The review feedback suggests refactoring the repeated score comparison into a boolean variable to improve readability and restoring explanatory comments for the counter-intuitive status field logic.

[gemini-code-assist]

条件 response_model.score == 1 在此处被重复使用了两次。建议将其提取为一个布尔变量（如 is_good），这样可以提高代码的可读性并遵循 DRY（Don't Repeat Yourself）原则。此外，建议保留关于 status 字段含义的注释，因为 status=False 表示“良好”这种逻辑比较反直觉，注释有助于后续维护。这符合 PEP 8 关于代码清晰度的建议。

Suggested change

	result = EvalDetail(
	metric=cls.__name__,
	status=not is_good, # True if Bad (issues found), False if Good
	status=False if response_model.score == 1 else True,
	score=response_model.score,
	label=["QUALITY_GOOD"] if is_good else [f"{response_model.type}.{response_model.name}"],
	label=["QUALITY_GOOD"] if response_model.score == 1 else [f"{response_model.type}.{response_model.name}"],
	is_good = response_model.score == 1
	result = EvalDetail(
	metric=cls.__name__,
	status=not is_good, # False 表示质量良好，True 表示存在问题
	score=response_model.score,
	label=["QUALITY_GOOD"] if is_good else [f"{response_model.type}.{response_model.name}"],