Add comparison for Xarray folks

[VeckoTheGecko]

Adding a comparison here to discuss with the Xarray folks at some point. See README

Keen to chat about this @fluidnumericsJoe in our meeting today

cc @erikvansebille

[VeckoTheGecko]

Was just exploring another avenue - what if we used a different backend (i.e., "cubed" instead of "dask") to see what that performance looks like.

I ran into this error out of the box

python xarray-profile.py
Traceback (most recent call last):
...
  File "/Users/Hodgs004/coding/repos/xarray-interpolation/xarray-profile.py", line 187, in run
    ds.isel(positions).compute()
    ~~~~~~~^^^^^^^^^^^
  File "/Users/Hodgs004/coding/repos/xarray-interpolation/.pixi/envs/default/lib/python3.14/site-packages/xarray/core/dataset.py", line 2867, in isel
    return self._isel_fancy(indexers, drop=drop, missing_dims=missing_dims)
           ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/Hodgs004/coding/repos/xarray-interpolation/.pixi/envs/default/lib/python3.14/site-packages/xarray/core/dataset.py", line 2923, in _isel_fancy
    new_var = var.isel(indexers=var_indexers)
  File "/Users/Hodgs004/coding/repos/xarray-interpolation/.pixi/envs/default/lib/python3.14/site-packages/xarray/core/variable.py", line 1141, in isel
    return self[key]
           ~~~~^^^^^
  File "/Users/Hodgs004/coding/repos/xarray-interpolation/.pixi/envs/default/lib/python3.14/site-packages/xarray/core/variable.py", line 831, in __getitem__
    data = indexing.apply_indexer(indexable, indexer)
  File "/Users/Hodgs004/coding/repos/xarray-interpolation/.pixi/envs/default/lib/python3.14/site-packages/xarray/core/indexing.py", line 1182, in apply_indexer
    return indexable.vindex[indexer]
           ~~~~~~~~~~~~~~~~^^^^^^^^^
  File "/Users/Hodgs004/coding/repos/xarray-interpolation/.pixi/envs/default/lib/python3.14/site-packages/xarray/core/indexing.py", line 467, in __getitem__
    return self.getter(key)
           ~~~~~~~~~~~^^^^^
  File "/Users/Hodgs004/coding/repos/xarray-interpolation/.pixi/envs/default/lib/python3.14/site-packages/xarray/core/indexing.py", line 1783, in _vindex_get
    raise TypeError("Vectorized indexing is not supported")
TypeError: Vectorized indexing is not supported

due to these lines https://github.com/pydata/xarray/blob/a1408234d1dc786965bfb6311f8782561ee0e676/xarray/core/indexing.py#L1782-L1783 . Seems that vectorized indexing is relatively new to the array API (data-apis/array-api#669 , data-apis/array-api#900) but there are still some limitations that (I think) prevent us from doing vectorized indexing - hopefully Stephan or someone can shed more light ehre.

Anyway - extracting the cubed array, trying the following (as expected) doesn't work either:

data = ds.U_grid_data.data
data[[2,3],[2,3],[2,3],[2,3]]
*** NotImplementedError: Only one integer array index is allowed.