Port XarrayZarrRecipe.subset_inputs to beam-refactor

[derekocallaghan]

XarrayZarrRecipe provides a subset_inputs feature:

pangeo-forge-recipes/pangeo_forge_recipes/recipes/xarray_zarr.py

Lines 762 to 765 in 45cc115

	:param subset_inputs: If set, break each input file up into multiple chunks
	along dimension according to the specified mapping. For example,
	``{'time': 5}`` would split each input file into 5 chunks along the
	time dimension. Multiple dimensions are allowed.

This is currently used by the NARR: Subsetting and OPeNDAP tutorial notebook. I checked the beam-refactor branch but I don't see an equivalent, where the only mention of subsetting is related to rechunking.

If it's the case that we no longer want to support this feature, this issue can be closed.

[rabernat]

We no longer need this feature. The beam refactor automatically handles sub-sampling into the target chunks.