Schema parser in Scala

[da-tubi]

closes #82

Fact

There are two places a Liga schema is written:

1. Python

   class Classifier(SklearnModelType):
   """Classification model type"""
   
   def schema(self) -> str:
       return "int"
   
   def predict(self, *args: Any, **kwargs: Any) -> List[int]:
       assert self.model is not None
       assert len(args) == 1
       return self.model.predict(args[0]).tolist()

   https://github.com/liga-ai/liga/blob/v0.2.0/python/liga/sklearn/models/classifier.py#L24-L33

2. SQL

   CREATE OR REPLACE MODEL schema_test
   RETURNS int
   USING 'mlflow:///{registered_model_name}';

Impl Change

Use parse_schema in Scala instead of in Python