Missing info regarding default compression behavior

[jm9176]

In MATLAB, no compression is applied if a user uses a default syntax, e.g. zarrcreate('prt_grp/arr1',[10 10]). However, if I create a zarr array using Python and Zarr library then it uses blosc compression by default. So, there seems to be some difference in default behavior between MATLAB and Python workflow.

So, it would be helpful to clarify this behavior or related differences.

**Python code**
# Create Zarr v2 array
import zarr
import numpy as np
store = zarr.DirectoryStore('prt_grp')
group = zarr.group(store=store,overwrite=True)
group.attrs['group_description'] = 'This is a sample Zarr group'
group.attrs['group_level'] = 1
array = group.create_dataset('arr_1',shape=(20,25),dtype=np.float32,chunks=(2,5),fillvalue=-9)
array.attrs['array_description'] = 'This is a sample Zarr array'
array.attrs['array_type'] = 'double'
array.attrs['array_level'] = 1
array[:] = np.random.rand(20,25)


**zarrinfo output from MATLAB**
>> info = zarrinfo('prt_grp/arr_1');
>> info.compressor

ans = 

  struct with fields:

    blocksize: 0
       clevel: 5
        cname: 'lz4'
           id: 'blosc'
      shuffle: 1

[abhibaruah]

The no compression default for MATLAB is by design (similar to our existing sci data functions).
We do not guarantee same behavior as the Python libraries.
With the PR #75
we will hopefully document this at the right location.

[abhibaruah]

Closing this since #75 has been merged.