Automl-alex

Fix
- same Fixses in AutoML
ADD
- New info in Readme.md

A big update that changes the logic of work
NEW
- Now processing the dataset is separated from the model for ease of use when you want to process the dataset yourself
- Separate transform allows us to save and transfer processing to new data
ADD
- Save & Load processing
- Save & Load model
- Reduce memory usage processing
- Detect and remove outliers

Fix
- score_cv_folds fix in ModelsReview
- normalization

ADD
- multivariate TPE sampler. This algorithm captures dependencies among hyperparameters better than the previous algorithm

Fix
- "ValueError non-broadcastable output operand..." in AutoMLRegressor

Fix
- DataConversionWarning in sklearn_models model.fit(X_train, y_train,)

Fix
- verbose in LinearRegression