Add support for HDFS storage

[Jimvin]

The hive-operator allows you to specify an S3 location but not a HDFS cluster. Add support to allow HDFS to be used as the storage target for hive tables.

Afterwards https://github.com/stackabletech/trino-operator/blob/bc442aa2f785208db5dd4ca98c2ba6d880370771/tests/templates/kuttl/smoke/check-s3.py#L91-L99 can be updated to the new mechanism.
It also servers as a perfect testcase

CRD proposal which matches the way we do in Trino: https://github.com/stackabletech/trino-operator/blob/bc442aa2f785208db5dd4ca98c2ba6d880370771/tests/templates/kuttl/smoke/10-install-trino.yaml.j2#L43-L57

spec:
  version: {{ test_scenario['values']['hive'] }}
  s3:
    reference: minio
  hdfs:
    configMap: hdfs

• Implemented
• Trino tests adopted (see comment above) and pass
• Feature tracker updated

[vsupalov]

Looks good 👍 I don't have a good idea of the usage though. There are three questions on my mind:

• Is only one hdfs or s3 entry possible? Could we have a list of each?
• Is it an "either", or a "both" when it comes to s3 and hdfs?
• Could it make sense to put these two into a parent, like trino does with "connector"?

If none of these are relevant, feel free to ignore them and move the ticket to the right.

[sbernauer]

Thx @vsupalov! To put it short

1. A list in general is not supported
2. Both can be specified at the same time
3. I don't think this makes sense, as hive has no concept for catalogs as Trino has