In the regression setting, cv=LeaveOneGroupOut() and cv=LeavePGroupsOut() are not working

[Valentin-Laurent]

Describe the bug
In the regression setting, cv=LeaveOneGroupOut() and cv=LeavePGroupsOut() are not working

To Reproduce
Script to reproduce the behavior:

from mapie.regression import MapieRegressor
from sklearn.datasets import load_iris
from sklearn.model_selection import *

X, y = load_iris(return_X_y=True)
groups = [0]*30 + [1]*30 + [2]*30 + [3]*30 + [4]*30

working = MapieRegressor(cv=GroupKFold())
working.fit(X, y, groups=groups)

not_working = MapieRegressor(cv=LeaveOneGroupOut()) # Same issue with LeavePGroupsOut(n_groups=n)
not_working.fit(X, y, groups=groups)

Additional context
The bug seems to come from this line of code, where check_no_agg_cv is not called with groups
self.use_split_method_ = check_no_agg_cv(X, self.cv, self.no_agg_cv_)

Simply adding groups here may solve the issue