A piezoelectric system would have faster responses than moving weights.

It's not an either-or proposition. As you burn propellant during course correction maneuvers, your center of gravity is going to change which will affect your axis of rotation. You could end up with some pretty uncomfortable wobbles. At the same time, you've got to be concerned with vibrations in a flexible boom. A truly rigid boom would be pretty heavy, so there would probably be some vibrations. In a worst case scenario, these vibrations could hit a harmonic and tear the boom apart. Not cool.

In a rigid-truss dumbbell ship, ideally main thrusters would be in the middle of the truss, maybe even counterspun. Timing spinning thrusters would not be difficult.

On spin stabilized satellites, delta-v maneuvers involve firing the thrusters in very short pulses as the vehicle rotates. Say that the thruster begins firing a short distance before the desired direction, say 15 degrees. It continues firing for an equal distance after the desired direction. The thruster is only pointing in the optimum direction for a very short time, wasting propellant. On a manned spacecraft, your rotation rate would be less than for a spin stabilized satellite, but the technique remains the same if you use thruster pulsing. The problems remain the same, too.

Putting the thrusters at the center of the boom can help in some respects but could also lead to some unpleasant vibration characteristics as well as some long propellant lines unless the propellant is in the center of the truss. Counterspinning across propellant lines wouldn't be easy.

The more stuff you have, the more stuff there is to break. You'll almost never go wrong if you follow the old engineering dictate, "Simplificate and build in lightness."

Your original idea was better - send the consumables and other heavy stuff by the slow boat and send the crew across as fast as possible. NASA is experimenting with a small diameter centrifuge to negate the effects of weightlessness. Aviation Week had a good article on this project a couple months ago. That would be a much better technique than spinning the whole spacecraft.