Functools.partial does not work properly for node outputs

[hunse]

Describe the bug

When using a functools.partial instance as a node output, we internally check inspect.isroutine on the output and if this is not true, we subtract 1 from the length of the argument list because we assume the first argument is self. However, functools.partial instances apparently do not register as routines, but do not have a self argument.

To reproduce

A minimal code snippet to reproduce the behavior:

import nengo

def input_fn(t, q):
    return t + q

with nengo.Network() as net:
    u = nengo.Node(partial(input_fn, q=3))

with nengo.Simulator(net) as sim:
    pass

Expected behavior

This should work. (e.g. should be able to probe node u above and get t + 3 as the output.

Error message and logs

/data/eric/workspace/nengo/nengo/node.py in coerce(self, node, output)
     71                 node.size_out = output.default_size_out
     72         elif callable(output):
---> 73             self.check_callable_args_list(node, output)
     74             # We trust user's size_out if set, because calling output
     75             # may have unintended consequences (e.g., network communication)

/data/eric/workspace/nengo/nengo/node.py in check_callable_args_list(self, node, output)
    134 
    135             if not required_len <= expected_len <= args_len:
--> 136                 raise self._fn_args_validation_error(output, self.name, node)
    137 
    138 

ValidationError: Node.output: output function 'functools.partial(<function input_f at 0x7f1c92addbf8>, phase_offset=0)' is expected to accept exactly 1 argument (time, as a float)

[drasmuss]

Ran into this same issue for a different reason in NengoDL, which I fixed like this https://github.com/nengo/nengo-dl/blob/fbf57c612668bfcc231788b0ed5dccd22695e271/nengo_dl/tensor_graph.py#L701. Should probably be able to do something similar.