[python] replace numpy.zeros with numpy.empty for the speedup

[StrikerRUS]

empty, unlike zeros, does not set the array values to zero, and may therefore be marginally faster. On the other hand, it requires the user to manually set all the values in the array, and should be used with caution.
https://numpy.org/doc/stable/reference/generated/numpy.empty.html

[jameslamb]

Makes sense to me!

I checked the blame and np.empty() has been a part of numpy for 15 years, so I don't think this introduces any compatibility problems: https://github.com/numpy/numpy/blame/623bc1fae1d47df24e7f1e29321d0c0ba2771ce0/numpy/matlib.py

[github-actions]

This pull request has been automatically locked since there has not been any recent activity since it was closed. To start a new related discussion, open a new issue at https://github.com/microsoft/LightGBM/issues including a reference to this.