ENH: sparse: Add indexing for 1D arrays

[dschult]

Adds support for indexing with 1D arrays.

Builds on #19833 (csr-1d) to provide indexing for 1D CSR arrays, though this code also works for DOK with minor changes after #19715 (dok-1d) gets merged.

The diff will be much easier to read after #19833 is merged. Look at only the last commit to see the diff for just this PR. I will rebase as needed, but I think the CSR PR will be merged soon and the diff here will become cleaner.

• adds test_indexing1d.py
• refactors IndexMixin._validate_indices to use an ndim-independent approach for both 1d and 2d.
• the supported formats will need methods _get_int, _get_slice and _get_array. These are dispatched to after processing the index.
• this tries to incorporate all improvements from the recently merged defect: sparse: 1d bool mask with wrong shape should raise IndexError #19957 (and doesn't change those tests).
• the ndim-independent approach means some helper functions are no longer needed and/or have been inlined. The two remaining helper functions are now methods so they have access to self.ndim. They are _as_indices and _compatible_boolean_array.

[dschult]

I've changed the tests to use np.testing assert_equal and assert_allclose in these new tests. I'll create a separate PR to change the new-but-merged-tests to use them too.

I've also added indexing support for np.newaxis/None (though it can't make 3D+). And that was an excuse to revamp some of the getitem/setitem code to be easier to move to nd. Quanstide's ndindex library would give almost everything we need for validate_indices, but it doesn't support boolean sparse arrays as indexes. Anyway, we can now index 2d and 1d sparse arrays.

I believe the test failures are not related to this PR.

Note: This PR does not change reduction operations to have them return 1D sparse arrays. That will be a separate PR.

Note: Another separate PR: We need to decide how to handle 0D. A[3,4] should return a 0D object as per array_api. But should it return a numpy or scipy.sparse 0D object? In general, an array type "should" return it's own type when indexed. But sparse is not a standard array type -- we don't expect the entire array api to be implemented. But the answer to this question will impact this same code.