Improve speed of train/selection/validation split function

[JanBenisek]

Currently, the funciton is using train_test_split() from sklearn twice, but with large datasets, the functions becomes slow and memory demanding due to the fact that we are creating multiple dataframes.
The solution would be just to append a list with the split [train, selection, train, validation ... ]

cobra/cobra/preprocessing/preprocessor.py

Line 340 in 9141313

def train_selection_validation_split(data: pd.DataFrame,