Preliminary Zarr Implementation + Tests

[MrFok]

• Generated separate files for Zarr implementation (without disturbing existing pipeline)
• Local test generated for Lazy Load Zarr Aggregation (LLZA)
• Minor dependency changes, bug fixes (needs checking), commenting
• Preliminary methods for multiple parts in collection-manager and granule-ingester

Changes in Proposed Pipeline

• Dataset metadata push for new, YAML format (needs dynamic public bucket boolean implementation)
• Streamlined processors for xarray Dataset instead of nexus processing
• Direct Lazy Load Zarr Aggregation write into S3 bucket instead of tile storage in Cassandra

NOTE: All changes have NOT been completely tested in a full run of the entire pipeline.