Guard against infinite loops - add timeout somehow if a test gets stuck?

Specific example: https://github.com/Terasology/SimpleFarming/issues/97

One theory is something goes weird and the MTE test just keeps generating chunks forever, which prints one of those log snippets about compacting the chunk cache every tick, which might be the only indication that something is still happening, otherwise it would just appear to hang.

I've seen similar cases elsewhere. For instance http://jenkins.terasology.io/teraorg/job/Terasology/job/engine/view/change-requests/job/PR-4036/ related to https://github.com/MovingBlocks/Terasology/pull/4036 for some reason got stuck several times for hours on end. One example printed this a bunch (although not nearly as much as the SimpleFarming PR):

11:50:39.535 [Test worker] ERROR o.t.w.b.internal.BlockManagerImpl - Attempt to fetch block with unknown uri 'engine:air'

It may not be an issue specific to the MTE, could be some underlying issue with environment-based tests. Not sure which actual test stalls for SimpleFarming, although the engine one is StorageManagerTest extends TerasologyTestingEnvironment so yeah not MTE, but maybe both hit that environment class?

Start by reproducing the SimpleFarming issue 97 case and inspect StorageManagerTest and TerasologyTestingEnvironment. Compare the behavior with the Jenkins PR-4036 run to determine whether both stalls share the environment-test path. Done means stalled tests reliably time out or otherwise report a clear failure instead of running indefinitely.