Add ARCO-OCEAN dataset to the open-data-registry

[stefanocampanella]

Description
This PR adds the ARCO-OCEAN dataset to the AWS Open Data Registry.

The ARCO-OCEAN dataset is a comprehensive analysis-ready, cloud-optimized dataset of physical properties of the ocean, waves, and sea-ice, along with hydrological and atmospheric forcing variables — designed to support machine-learning forecasting of the ocean state.

Key Features:

• Covers ~28 years of data (from 1 January 1993 to 30 June 2021) including ocean circulation, sea-ice, waves, atmospheric and hydrological variables.
• Provided on a regular 0.25° horizontal grid with 1 day temporal resolution, and 10 vertical levels for ocean variables.
• Distributed in the Zarr format (v2) with consolidated metadata, chunked by time (1 day) and compressed (LZ4).
• Documentation and tutorials available at the GitHub repository: https://github.com/inogs/arco-ocean.
• Released under the CC-BY license (Creative Commons Attribution) so users are free to reuse with attribution.

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.