Fix creating tensor from copy of numpy array warning messages

[arnavgarg1]

This PR is created for the following issue.

PyTorch wants us to create a tensor for input and output features from a copy of the numpy array instead of directly using the numpy array. This typically results in the following error message:

UserWarning: The given NumPy array is not writable, and PyTorch does not support non-writable tensors. This means writing to this tensor will result in undefined behavior. You may want to copy the array to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at  /root/pytorch/torch/csrc/utils/tensor_numpy.cpp:172.)

This PR should fix this everywhere within Ludwig.

[github-actions]

Unit Test Results

       6 files  ±0         6 suites  ±0   2h 13m 50s ⏱️ - 9m 1s
2 828 tests ±0  2 794 ✔️ ±0    34 💤 ±0  0 ❌ ±0 
8 484 runs  ±0  8 378 ✔️ ±0  106 💤 ±0  0 ❌ ±0 

Results for commit ddc1c74. ± Comparison against base commit af4b102.

♻️ This comment has been updated with latest results.