SIFS may not shut down properly with compactor still running

The SoftIndexFileStoreStressTest can fail to shutdown which seems to be related to the compactor running when shutdown is initiated. This can cause the server to be force killed which will require a full index rebuild which will cause the server to take much longer to restart next time. It needs to be investigated if this is caused by an exception that isn't logged or if the compactor actually completed but didn't notify properly.