Slowrotor
Scaling is a difficult engineering aspect as major elements such as lenght, weight and strenght don't scale equally. As a result, different size ranges will require different architectural solution.
I assume you are talking a small model.
In this range I am pretty sure 1 rotor will be aerodynamically optimal.
Next comes the electrical motors, again I think lesser will be better.
You raised a second and in may opinion different design criterium about controlability and redundancy.
I think that the first one can be equally realised by some chip governing say 4 servos or a heli kind of cyclic coupled with a gyro.
Redancy is a whole different matter. I have the feeling that even in the real world not everybody is align on the question single or bi-engine.
m2cents, d3