Add WaitingForRelease feature flag

[sebrandon1]

As of right now, when we are adding new tests to the repo we can't really release a new version of the certsuite until everything has been merged in. This will allow us to set tests as disabled and they won't affect existing release version of the certsuite.

Changes:

• Adds a WaitingForRelease tag similar to TagCommon, TagTelco, etc.
• v1.6 tests that have been newly added have had WaitingFoRelease added to them. This removes their entries from the CATALOG.md temporarily until we are ready to fully release them to partners.

To keep the tests enabled in our CI's we are setting an environment variable REDHAT_CI which will always set the test cases to enabled regardless of `waiting-for-release" tag.

[dcibot]

from change #2614:

• SUCCESS https://www.distributed-ci.io/jobs/0f9d5aaa-57e3-4be7-b2b7-59c76358fd3a/jobStates

[dcibot]

from change #2614:

• FAILURE https://www.distributed-ci.io/jobs/0f875783-0873-43de-892e-d39138d88513/jobStates

[sebrandon1]

This is probably going to fail DCI until they add the environment variable. Right now I have it set as REDHAT_CI but I could be persuaded to use another variable or way to differentiate how to add the checks into the group that are waiting for release.

This environment variable (whatever we decide on) will probably also have to be set on developer machines who are attempting to run the latest code.

[dcibot]

from change #2614:

• SUCCESS https://www.distributed-ci.io/jobs/f0df1ba6-4862-4ca4-b211-7f0cf30bdfc0/jobStates

[dcibot]

from change #2614:

• SUCCESS https://www.distributed-ci.io/jobs/722cf774-b021-4e46-8809-ba60f247b43e/jobStates

[greyerof]

Just my two cents here: if a test case made it (after the review+approval process) to the main branch, it is then supposed to be fully functional and, at least, it should have been manually tested, so appearing in the catalog as a new official test case in the next release cut seems 100% legit to me in that case. Why should we hide them and wait for some X.Y.Z release to announce all of them at once?

In case we want/need to go with half-baked test cases, why not just flag them with some extra label like that "waiting-for-release" or similar ("release-pending", "not-released", "in-progress"...)? It should be sufficient to state in our official documentation that the results of those test cases, whether they pass or fail, should not be considered official, as they are not officially released/supported yet. Just removing that label when they're 100% ready or before the release cut should work. Adding extra env vars shouldn't be necessary.

[sebrandon1]

That's fair. There's nothing "wrong" with just releasing what we currently have merged into main other than just keeping it more organized for users. I agree the environment variable makes it ugly. I'll mark this as closed and we can discuss it more in the future if needed.

In the meantime, I think I will just release 5.4.x with the currently merged test cases. That's how we've been doing it in the past if there are any new test cases by just bumping the minor version. (major.minor.patch).