I'd imagine combining slow liftoffs with thrust vectoring, you wouldn't be looking at the type of motors normally used today in model rocketry. You'd be looking at high total impulse and low average thrust motors such as new AeroTech G8 (
https://www.thrustcurve.org/motorsearch.jsp?id=1122). 136Ns of impulse, but only only 8N of average thrust (peaks at only 15N). 16+ seconds burn time. Because you wouldn't need that high average thrust to get the rocket up to speed quickly before it leaves the rail as you need with passive fins-only stability control. If thrust vectoring becomes more widespread, I'd expect we'll see more motors with low thrust to total impulse ratio being developed.