[BUG] Unexpected keyword argument 'embedding_model' for passage filter module

[rjwharry]

Describe the bug
unexpected keyword argument 'embedding_model' error occurred in passage filter module such as SimilarityThresholdCutoff._pure()

To Reproduce

1. set embedding_model of similarity_threshold_cutoff in config.yaml
2. Run autorag

Expected behavior
Succeed

Full Error log
TypeError: SimilarityThresholdCutoff._pure() got an unexpected keyword argument 'embedding_model'

Code that bug is happened
After init cls instance, embedding_model key must be popped from kwargs. But kwargs in BaseModule and SimilarityThresholdCutoff have different id like deep copy.

class BaseModule(metaclass=ABCMeta):
    @classmethod
        def run_evaluator(
            cls,
            project_dir: Union[str, Path],
            previous_result: pd.DataFrame,
            *args,
            **kwargs,
        ):
            instance = cls(project_dir, *args, **kwargs)
            result = instance.pure(previous_result, *args, **kwargs)
            del instance
            return result
class SimilarityThresholdCutoff(BasePassageFilter):
    def __init__(self, project_dir: str, *args, **kwargs):
        super().__init__(project_dir, *args, **kwargs)
        embedding_model_str = kwargs.pop("embedding_model", "openai")
        self.embedding_model = embedding_models[embedding_model_str]()

[rjwharry]

I'm figuring out how to pass kwargs as shallow copy