Add a reliable "write to cloud storage" archive service implementation

[jsync-swirlds]

Create a new archive class service, which may deploy in addition to other archive service(s).
This archive service will, based on separate configuration, take archived packages of block files (ideally quite large files) and reliably migrate those files to cloud storage.

Suggested features

• Support for both GCP cloud storage and Amazon S3
• Direct API calls, without use of cloud provider SDKs
  • Handle and retry for all non-fatal errors.
• Verify-before-remove
  • calculate and verify a file hash (e.g. MD5 or SHA2) and verify against the hash calculated by the cloud storage provider.
  • Only remove the local file if the hash values match.
  • Retry if the hash values do not match.
• Upload on background Virtual Threads.