Trazodone decreases platelet serotonin only mildly (remember, it is fairly uniquely is a noncompetitive SERT inhibitor) so it shouldn't have much effect on function donation through that mechanism.
Mirtazapine does not change platelet serotonin levels, but it does block the 5HT2A receptor which is where that serotonin needs to bind on platelets. Notably, this mechanism is potentially synergistic with SRIs (one decreases the amount of serotonin available to be released, the other antagonizes the activity of that serotonin) which may be important to consider as they are frequently used together, but admittedly I haven't looked at any studies on the matter. I suspect mirtazapine's alpha-2 antagonism also contributes to risk through altering hemodynamics.
Bupropion should have no direct effect on platelet function and should not affect bruising. It may affect bleeding risk, but remember that is a function of primary and secondary coagulation function, while bruising is really only primary coagulation (i.e. dependent on platelet function. I think it is reasonable to trial bupropion if patients are having trouble with bruising.