Accessing unset f_args and jac_args attributes of ivp.IVP class should return None

[davidrpugh]

Current behavior raises an AttributeError. This is undesirable. I think fix, just requires changing the setter for the attributes as follows...

def f_args(self, new_f_args):
        """Set new values for the parameters."""
        if not isinstance(new_f_args, (tuple, type(None))):
            mesg = "IVP.args must be a tuple, not a {}."
            raise AttributeError(mesg.format(new_f_args.__class__))
        elif new_f_args is not None:
            self.set_f_params(*new_f_args)
            self._f_args = new_f_args
        else:
            pass

to

def f_args(self, new_f_args):
        """Set new values for the parameters."""
        if not isinstance(new_f_args, (tuple, type(None))):
            mesg = "IVP.args must be a tuple, not a {}."
            raise AttributeError(mesg.format(new_f_args.__class__))
        else:
            self.set_f_params(*new_f_args)
            self._f_args = new_f_args

Similar fix for the jac_args setter. PR for this will follow shortly.

[davidrpugh]

Actually, this issue has made me realize that I made a "poor" design decision in the way that I allow users to set arguments for the functions f and jac.

Currently there are two attributes of the ivp.IVP class that are serving this purpose: f_args and f_params (also jac_args and jac_params). I introduced the new f_args and jac_args attributes in order to conduct explicit type checking (something which scipy.integrate.ode should be doing but isn't).

However, I now think that the existence of these different attributes is likely to confuse users and I would like to remove both the f_args and jac_args attributes. This will require changes to the class, tests, and example notebook.

Thoughts?