When pull requests have RuboCop offenses they need to click
"Build and run RuboCop" section and scroll about 200 lines
to skip `bundle install` output.
This change splits "Build and run RuboCop" into two parts.
then "Run RuboCop" section only shows `bundle exec rubocop --parallel`
output
Follows up #37582
GitHub Actions cache requires `path:` parameter to configure
which directory to be cached. It looks there is no easy way to retrieve path information
using bundler command then workaround implemented to use `vendor/bundle`.
Thanks to https://github.com/actions/cache/pull/4
Updated the default Ruby workflow:
* Trigger this action for pull request and push
* Install requried OS packages for Active Record connection adapters
* Removing `gem install bundler` because Ruby 2.6 already installs it
* Run RuboCop in parallel by `bundle exec rubocop --parallel`
* Rename `ruby.yml` to `rubocop.yml` to represent the workflow
Bump RuboCop to 0.76.0 and rename `Style/UnneededPercentQ` to `Style/RedundantPercentQ`
```ruby
$ bundle exec rubocop
Error: The `Style/UnneededPercentQ` cop has been renamed to `Style/RedundantPercentQ`.
(obsolete configuration found in .rubocop.yml
```
Removed `.codeclimate.yml`
