Create a Zarr + virtual store publication policy

[abarciauskas-bgse]

Goals: We want to share the Zarr (native + virtual) datasets more broadly. Right now they have been produced and stored a bit adhoc. We need to be consistent about what bucket(s) they go in, what policies those buckets have and how they are catalogued.

Additional questions/considerations:

1. Where should the datasets be discoverable? There is Earthmover marketplace, the VEDA STAC catalog, NASA CMR and AWS PDS. AWS PDS has a special NASA page (https://registry.opendata.aws/collab/nasa/) but it looks like this includes some protected buckets. Who can we talk to about better curating this?
2. Publish Zarr + Virtual Zarr datasets to STAC #392

cc @maxrjones who mentioned this during our checkin meeting. Please add any important additional considerations I may have missed.