DOC Explicitly remove support for precompute='auto' for Lasso (scikit-learn#19412)

Author: jjerphan

sklearn/linear_model/_coordinate_descent.py

@@ -939,12 +939,11 @@ class Lasso(ElasticNet):
         :class:`~sklearn.preprocessing.StandardScaler` before calling ``fit``
         on an estimator with ``normalize=False``.
 
-    precompute : 'auto', bool or array-like of shape (n_features, n_features),\
+    precompute : bool or array-like of shape (n_features, n_features),\
                  default=False
         Whether to use a precomputed Gram matrix to speed up
-        calculations. If set to ``'auto'`` let us decide. The Gram
-        matrix can also be passed as argument. For sparse input
-        this option is always ``False`` to preserve sparsity.
+        calculations. The Gram matrix can also be passed as argument.
+        For sparse input this option is always ``False`` to preserve sparsity.
 
     copy_X : bool, default=True
         If ``True``, X will be copied; else, it may be overwritten.

sklearn/linear_model/tests/test_coordinate_descent.py

@@ -813,9 +813,11 @@ def test_precompute_invalid_argument():
         assert_raises_regex(ValueError, ".*should be.*True.*False.*auto.*"
                             "array-like.*Got 'invalid'", clf.fit, X, y)
 
-    # Precompute = 'auto' is not supported for ElasticNet
+    # Precompute = 'auto' is not supported for ElasticNet and Lasso
     assert_raises_regex(ValueError, ".*should be.*True.*False.*array-like.*"
                         "Got 'auto'", ElasticNet(precompute='auto').fit, X, y)
+    assert_raises_regex(ValueError, ".*should be.*True.*False.*array-like.*"
+                        "Got 'auto'", Lasso(precompute='auto').fit, X, y)
 
 
 def test_elasticnet_precompute_incorrect_gram():