Investigate and either fix or quell zarr.zip warnings

[JoerivanEngelen]

#1706 implements support for zarr and zarr.zip. The latter however gives quite some warnings about duplicate objects, which are indeed created:

There is some documentation on this: https://zarr.readthedocs.io/en/v2.15.0/api/storage.html#zarr.storage.ZipStore
We have to see if we can align chunks in some way to avoid this warning.

@Manangka also got the following warning which requires some investigation:

UnstableSpecificationWarning: The data type (FixedLengthUTF32(length=4, endianness='little')) does not have a Zarr V3 specification. That means that the representation of arrays saved with this data type may change without warning in a future version of Zarr Python. Arrays stored with this data type may be unreadable by other Zarr libraries. Use this data type at your own risk! Check https://github.com/zarr-developers/zarr-extensions/tree/main/data-types for the status of data type specifications for Zarr V3.
v3_unstable_dtype_warning(self)

[JoerivanEngelen]

I see there is a align_chunks argument, which might help:

https://docs.xarray.dev/en/stable/generated/[xarray.Dataset.to_zarr](https://docs.xarray.dev/en/stable/generated/xarray.Dataset.to_zarr.html).html

[JoerivanEngelen]

@Manangka also got the following warning which requires some investigation:

UnstableSpecificationWarning: The data type (FixedLengthUTF32(length=4, endianness='little')) does not have a Zarr V3 specification. That means that the representation of arrays saved with this data type may change without warning in a future version of Zarr Python. Arrays stored with this data type may be unreadable by other Zarr libraries. Use this data type at your own risk! Check https://github.com/zarr-developers/zarr-extensions/tree/main/data-types for the status of data type specifications for Zarr V3.
v3_unstable_dtype_warning(self)

To reproduce:

import xarray as xr
from pathlib import Path
import shutil

da = xr.DataArray(["a", "bb", "cccc"], dims="x", name="letters", coords={"x": [0, 1, 2]})
ds = da.to_dataset()
ds.to_netcdf("test.nc")
ds.close()

path = Path("test.zarr")
if path.exists():
    # Check if directory (ordinary .zarr, directory) or ZipStore (zip file).
    if path.is_dir():
        shutil.rmtree(path)
    else:
        path.unlink()

ds2 = xr.open_dataset("test.nc")
ds2.to_zarr("test.zarr")
ds2.close()

ds3 = xr.open_zarr("test.zarr")
ds3.close()

This for now can be quelled by setting ds2.to_zarr("test.zarr", zarr_format=2). That has the disadvantage that we cannot use the neat new features of zarr 3. I don't read anything about string encoding here. This mentions it needs to be variable length UTF-8, but I don't see a way to set that when calling to_zarr.