Write the numba implementation for models S-J 2D and 3D

Models Stefanescu-Jirsa 2D and 3D in TVB have more than one mode, and only a python defined dfun.

They could be rewritten with a numba dfun, considering the following aspects:

• other numba gufuncs drop the mode dimension, which isn't ok here, so the gufunc signature should have the extra mode dimension
• some derived parameters are matrices and need to have a (n_mode, n_mode) shape in the gufunc signature instead of (1,)
• these models have a dot product between matrix parameters along the mode dimension, might require tweaking

This could be a unit test:

 
class TestNumbaDfun:
    def dfun(self, *args, **kwargs):
        f1 = self._numpy_dfun(*args, **kwargs)
        f2 = super().dfun(*args, **kwargs)
        assert np.allclose(f1, f2)
        return f2
 
class TestModel(TestNumbaDfun, ReducedHindmarshRoseNumba):
    pass
 
sim = simulator.Simulator(model=TestModel(), ...)
sim.run()