Computes the p-norm distance between every pair of row vectors in the input.
This is identical to the upper triangular portion, excluding the diagonal, of
`torch_norm(input[:, None] - input, dim=2, p=p)`

. This function will be faster
if the rows are contiguous.

## Arguments

- input
input tensor of shape \(N \times M\).

- p
p value for the p-norm distance to calculate between each vector pair
\(\in [0, \infty]\).

## Details

If input has shape \(N \times M\) then the output will have shape
\(\frac{1}{2} N (N - 1)\).