Run Full QA Suite When Toggling Feature Flags in Staging

Problem

gitlab-com/chatops!376 (merged) was recently implemented to only run the smoke and reliable test suite when a feature flag is toggled for Staging.

Now that most of the scheduled runs have been deactivated, this was one of our last regular cadences of running the full QA suite in Staging. When this issue was first reported on 2023-06-14, we hadn't seen a full Staging QA run since 2023-06-09 (5 days of no full QA runs).

For Staging, we have several non-reliable QA tests that can only be run on that environment. An example of this is the Fulfillment QA tests, as it relies on CustomersDot and Zuora Central sandbox set up there. These tests currently can't be converted to reliable as it heavily relies on Zuora integration, which is not very reliable due to callout delays, outages, etc. They also can't run against production, since it works with credit card purchases and also needs an admin account for some of the test setup.

Developers from various groups also rely on the results from a full test run when toggling feature flags (ex: groupthreat insights in GitLab.org / Quality Department / Engineering Productivity / team and grouppackage registry , see comment here in the #quality Slack channel as well)

Proposal

Please consider re-implementing triggering the full QA suite when a feature flag is toggled in Staging