[Kernel] refactored table_property_enablement_key (#5245)

[ju-klein]

Which Delta project/connector is this regarding?

• Spark
• Standalone
• Flink
• Kernel
• Other (fill in here)

Description

This PR adds #5245.

Rationale:

• TablePropertyEnablementKey should exist for every TableFeature.
• We are alllowed to access SET_TABLE_FEATURE_SUPPORTED_PREFIX from TableFeature since it is also accessed from other places.

How was this patch tested?

Since this does not add new functionality, I just ran the test suite, which passed.

Does this PR introduce any user-facing changes?

No

[ju-klein]

I did install the format checker, it just didn't run. I did not have reformat in "actions on save".
I'll fix it asap.

[ju-klein]

Sorry guys, at this moment, neither do I know why the formatting would be off, nor which file still needs to be formatted.
I installed the format checker, I ran it manually as well as automatically when saving files. o0
I'll look at this tomorrow.

[scottsand-db]

@ju-klein can you please run build/sbt kernelGroup:javafmtAll && build/sbt kernelGroup:scalafmtAll

[ju-klein]

I was able to fix the formatting issues. Five checks fail.
Most of the checks seem to fail at tests which involve io.delta.kernel.exceptions.KernelException: Unsupported Delta table feature: table requires feature "domain_metadata_w_feature" which is unsupported by this version of Delta Kernel..

Only one check fails at buildCreateTableTransaction sets required properties and uses UC committer *** FAILED *** [info] io.delta.kernel.exceptions.KernelException: Unsupported Delta table feature: table requires feature "catalog_managed_r_w_feature_preview" which is unsupported by this version of Delta Kernel..

So far, my investigation has shown that DOMAIN_METADATA_W_FEATURE is a DomainMetadataFeature() which extends TableFeature.WriterFeature which extends TableFeature, which has the getTablePropertyElementKey method.

I'll investigate further.

[ju-klein]

Do the CI/CD checks include the test suite?
If so, this can be merged.
Also: Should I try to squash the commits before?
Edit: I don't merge this myself, do I? o0

[ju-klein]

Hi @scottsand-db ,
sorry to ping you directly. I am not sure how to proceed. Should I myself update the branch (do I even have permissions for that) or does someone of you guys do that? Should I not have requested another review? I don't want to spam you guys, I am just unsure about the proper protocol here. ^^'

[scottsand-db]

LGTM!