We have an issue with our rules setup that creates a fairly big inefficiency. In the case of `trigger-omnibus` build failing, downstream pipeline `e2e:package-and-test` is still being triggered due to current rules setup where `trigger-omnibus` is allowed to fail in all cases but `dependency changes`. This results in a rather big waste as e2e downstream pipeline is triggering 50+ test jobs which all fail due to docker image not existing in registry. A separate issue is with [gitlab-qa](https://gitlab.com/gitlab-org/gitlab-qa) which due to retry setup takes a long time to fail. Example pipeline: https://gitlab.com/gitlab-org/gitlab/-/pipelines/1237792605 <!-- template sourced from https://gitlab.com/gitlab-org/gitlab/-/blob/master/.gitlab/issue_templates/Default.md -->