chore: Prepare for 0.7.0 development

[andygrove]

Which issue does this PR close?

Closes #.

Rationale for this change

What changes are included in this PR?

How are these changes tested?

[andygrove]

How was this even working? Perhaps this is unused?

[andygrove]

Ah, there was a corresponding incorrect version in setup-spark-builder:

  comet-version:
    description: 'The Comet version to use for Spark'
    required: true
    default: '0.5.0-SNAPSHOT'

[andygrove]

We also set the comet version dynamically:

../mvnw -nsu -q versions:set-property -Dproperty=comet.version  -DnewVersion=${{inputs.comet-version}} -DgenerateBackupPoms=false

[kazuyukitanimura]

It is perhaps the best to separate the PR for this 0.6.0 fix otherwise, branch-0.6 is left in a wrong state

[andygrove]

I filed #1405 to fix the issue in branch-0.6

[andygrove]

@kazuyukitanimura I pushed another commit to remove all of the hard-coded Comet version number from GitHub workflows/actions.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 39.32%. Comparing base (f09f8af) to head (1e11c14).
Report is 37 commits behind head on main.

Additional details and impacted files

@@              Coverage Diff              @@
##               main    #1404       +/-   ##
=============================================
- Coverage     56.12%   39.32%   -16.81%     
- Complexity      976     2081     +1105     
=============================================
  Files           119      265      +146     
  Lines         11743    61132    +49389     
  Branches       2251    12962    +10711     
=============================================
+ Hits           6591    24040    +17449     
- Misses         4012    32587    +28575     
- Partials       1140     4505     +3365     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[andygrove]

@kazuyukitanimura I made more changes since you approved - could you take another look?

[andygrove]

This was hiding the fact that we had an incorrect version in the diff files. Now that this is removed, CI will fail if we do not have the correct version in the diff files, which is good since we want depend on these files having the correct version when running the Spark SQL tests locally.

[kazuyukitanimura]

loks good

[andygrove]

Thanks @kazuyukitanimura