DOC Remove outdated comments regarding memoryviews (scikit-learn#27169)

Signed-off-by: Julien Jerphanion <git@jjerphan.xyz>

Author: jjerphan

sklearn/cluster/_k_means_common.pyx

@@ -4,10 +4,6 @@
 #
 # License: BSD 3 clause
 
-# TODO: We still need to use ndarrays instead of typed memoryviews when using
-# fused types and when the array may be read-only (for instance when it's
-# provided by the user). This is fixed in cython > 0.3.
-
 import numpy as np
 from cython cimport floating
 from cython.parallel cimport prange

sklearn/cluster/_k_means_elkan.pyx

@@ -2,10 +2,6 @@
 #
 # Licence: BSD 3 clause
 
-# TODO: We still need to use ndarrays instead of typed memoryviews when using
-# fused types and when the array may be read-only (for instance when it's
-# provided by the user). This is fixed in cython > 0.3.
-
 from cython cimport floating
 from cython.parallel import prange, parallel
 from libc.stdlib cimport calloc, free

sklearn/cluster/_k_means_lloyd.pyx

@@ -1,9 +1,5 @@
 # Licence: BSD 3 clause
 
-# TODO: We still need to use ndarrays instead of typed memoryviews when using
-# fused types and when the array may be read-only (for instance when it's
-# provided by the user). This is fixed in cython > 0.3.
-
 from cython cimport floating
 from cython.parallel import prange, parallel
 from libc.stdlib cimport malloc, calloc, free

sklearn/cluster/_k_means_minibatch.pyx

@@ -1,7 +1,3 @@
-# TODO: We still need to use ndarrays instead of typed memoryviews when using
-# fused types and when the array may be read-only (for instance when it's
-# provided by the user). This will be fixed in cython >= 0.3.
-
 from cython cimport floating
 from cython.parallel cimport parallel, prange
 from libc.stdlib cimport malloc, free