Why, why do you not think about these things and offer an option to just restart or end the push back anyway?
You are assuming we haven't "thought" about that, when in fact it's exactly the opposite, and the pushback has TWO different ways to end: Abort and End. What's happening here, is something completely different:
the plane jumps and the pushback stops the Tug disappears
This seems to indicate the scenery had a bump/slope so big, that might have caused the flight model to "fly" for a second, hence the jump, and this might have caused the Tug to disconnect, because GSX assumed you flew away intentionally.
Them there's another issue, which instead is the "NW STEER DISENG", which I GUESS ( I can't possibly know, but it's likely the case, considering how the Fenix works ), it's a Fenix custom system so, it's not something we could control, even in case we had to remove the tug to because of that abnormal jump.
The only thing we can do, is checking if, when this happens, all the various variables published by GSX for aircraft developers to recognize GSX current state, are correctly updated even in that particular case, so they can possibly update all their affected systems, which we don't have any way to control.
I edited this thread title with a more appropriate description of the problem because, "I'm sick of this", won't help any other user finding it in the future.