Update dependency numpy

[samir-nasibli]

Description

Update numpy dependency version. Currently avoiding using numpy 2.0 as a workaround. The problem should be fixed on the sklearnex side.

Numpy 2.0 in env causes:

   File "/sources/runner.py", line 25, in <module>
     import utils
   File "/sources/utils.py", line 25, in <module>
     from datasets.make_datasets import try_gen_dataset
   File "/sources/datasets/make_datasets.py", line 21, in <module>
     from sklearn.datasets import make_classification, make_regression, make_blobs
   File "/env/lib/python3.10/site-packages/sklearn/__init__.py", line 82, in <module>
     from .base import clone
   File "/env/lib/python3.10/site-packages/sklearn/base.py", line 17, in <module>
     from .utils import _IS_32BIT
   File "/env/lib/python3.10/site-packages/sklearn/utils/__init__.py", line 19, in <module>
     from .murmurhash import murmurhash3_32
   File "sklearn/utils/murmurhash.pyx", line 1, in init sklearn.utils.murmurhash
 ValueError: numpy.dtype size changed, may indicate binary incompatibility. Expected 96 from C header, got 88 from PyObject

[ethanglaser]

It may be preferable to upgrade sklearn version instead

[samir-nasibli]

It may be preferable to upgrade sklearn version instead

It was created before numpy bump up on sklearnex side, so now make sense update sklearn version. I was using this branch for my measurements. I think using not numpy 2.0 for latest skl versions also valid case for now. I would like now go first with this changes, and jut update the condition on https://github.com/IntelPython/scikit-learn_bench/pull/151

[Alexsandruss]

Numpy is not explicitly specified as dependency after refactoring