Make fixtures independent from production data

What does this MR do and why?

This reduces fixtures to dummy data so that we test the logic instead of the data.

For the few tests that rely on specific groups/stages value, we stub the files with the minimal required amount of data.

This removes the need to regularly update the fixtures from about.gitlab.com, and remove the need to make any changes to spec/lib/devops_labels_spec.rb as a consequence of these updates.

This also removes a lot of useless duplicated data (e.g. 45k lines from team.yml).

This doesn't remove the need to regenerate CI jobs and policies upon group/stage/section renaming (but we should probably not run the regeneration-check in merge requests).

Action items

• If adding environment variables for reactive processors, update config/triage-web.yaml and .gitlab/ci/triage-web.yml
• (If applicable) Add documentation to the handbook pages for Triage Operations =>
• (If applicable) Identify the affected groups and how to communicate to them:
  • /cc @person_or_group =>
  • Relevant Slack channels =>
  • Engineering week-in-review