TST: Remove cloneable test for multi-shell kernel

Cloning is primarily intended for classes that are meant to be
estimators (i.e. inheriting from `BaseEstimator`).

Author: jhlegarreta

test/test_model_gpr.py

@@ -26,7 +26,6 @@
 import numpy as np
 import pytest
 from dipy.io import read_bvals_bvecs
-from sklearn.base import clone
 
 from nifreeze.model import gpr
 
@@ -589,12 +588,6 @@ def test_multishellkernel_nonpositive_bval(bad_bval, match):
         k(X)
 
 
-def test_multishellkernel_cloneable():
-    k = gpr.MultiShellKernel()
-    k2 = clone(k)
-    assert isinstance(k2, gpr.MultiShellKernel)
-
-
 @pytest.mark.parametrize("n_samples", [8])
 def test_multishellkernel_gp(n_samples):
     """Smoke test: DiffusionGPR can fit/predict with MultiShellKernel."""