[PyTorch] Fix in matmul function that enables working with all sizes …

[padreofthegame]

…of input tensors

This PR should enable full functionality of matmul operator in PyTorch frontend. In addition to current behavior, it introduces:

• matrix multiplication of two 1D tensors, 1D x 1D,
• matrix multiplication of 1D and ND tensors, 1D x ND.

It also improves operator functionality when working with:

• matrix multiplication of ND and 1D tensors, ND x 1D,
• matrix multiplication of two ND tensors, ND x ND.

Also, proper assertions are added.

[tvm-bot]

Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews from Reviewers by @-ing them in a comment.

• cc @shingjan, @yelite _{See #10317 for details}

_{Generated by tvm-bot}