Separate structures for Euler rotation orders

[EwoudSmeur]

As @tcunis pointed out, we may define more structures in pprz_algebra_float.h for each of the euler rotation orders that we use besides ZYX (like YXZ and ZXY), such that each function that takes euler angles as an argument can only take the right ones.

Opinions?

[kirkscheper]

You could also add a flag to the struct that defines the desired rotation order.

[EwoudSmeur]

But what do we do if a function does not support a certain order? Then that has to be dealt with at run-time, instead of preventing compilation.