Backport of #25675.

This PR adds new matvec and vecmat gufunc to calculate the matrix-vector and vector-matrix product, to add to plain matrix multiplication with matmul and the inner vector product with vecdot.

Fixes #12348

Note that for complex numbers, vecmat is defined as x†A, i.e., the complex conjugate of the vector is taken. This seems to be the standard and is what we used for vecdot too (x†x). However, it is not what matmul does for vector-matrix or indeed vector-vector products. I think this is a bug in matmul, which I'm happy to fix here. But I'll post to the mailing list to get feedback.

Separately, with these functions available, in principle these could be used in __matmul__ and the specializations in np.matmul removed. But that can be a separate PR (if it is wanted at all).

Summary of mailing list discussion

• Behaviour of np.matmul for vector-matrix should not be adjusted. Some surprise that "matrix multiplcation" deals with vectors at all.
• Less obvious consensus on what @ should do.
• Moderate support for having special matvec and vecmat functions.