Hi I think that LassoCV and RidgeCV classes do not have a way to disable refitting after finding the best hyperparameters. For more control over the refitting process, you can use GridSearchCV or RandomizedSearchCV with the refit=False.

from sklearn.model_selection import GridSearchCV
from sklearn.linear_model import Lasso

param_grid = {'alpha': [0.1, 1, 10]}
grid = GridSearchCV(Lasso(), param_grid, refit=False)
grid.fit(X, y)

best_params = grid.best_params_

Or you can do a monkey-patching which I usually do when something is not working as I expect.
I looked into this solution it seems a little bit complicated here.

Noted. Thank you @alifa98

I'll convert this into a feature request issue to allow disabling refitting.