Add support for tiling an array to the specification

[kgryte]

This PR

• resolves https://github.com/data-apis/array-api/issues/655 by adding an API for creating a new array by tiling an input array a specified number of repetitions.
• requires that the second positional argument (repetitions) be a tuple of ints. NumPy et al support providing an integer; however, PyTorch and TensorFlow do not. PyTorch requires a tuple. TensorFlow requires an array. This PR seeks to find common ground by requiring repetitions to be a tuple.
• supports prepend behavior, as supported by NumPy et al and PyTorch. I.e., if len(repetitions) is less than the rank of x, then ones are prepended to repetitions. If the rank of x is less than len(repetitions), singleton dimensions are prepended to x. TensorFlow does not support this behavior, requiring that repetitions.shape[0] equal the rank of x.

[kgryte]

As this PR has been discussed during Consortium meetings and no objections have been raised either here or in the original RFC (https://github.com/data-apis/array-api/issues/655), will merge. Any further changes can be addressed in follow-up PRs...