Some model specs are not json-serializable by CLI

[davemfish]

(C:\Users\dmf\projects\invest\env) λ invest getspec --json pollination
Traceback (most recent call last):
  File "C:\Users\dmf\projects\invest\env\lib\runpy.py", line 194, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "C:\Users\dmf\projects\invest\env\lib\runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "C:\Users\dmf\projects\invest\env\Scripts\invest.exe\__main__.py", line 7, in <module>
  File "C:\Users\dmf\projects\invest\env\lib\site-packages\natcap\invest\cli.py", line 420, in main
    message = json.dumps(spec)
  File "C:\Users\dmf\projects\invest\env\lib\json\__init__.py", line 231, in dumps
    return _default_encoder.encode(obj)
  File "C:\Users\dmf\projects\invest\env\lib\json\encoder.py", line 199, in encode
    chunks = self.iterencode(o, _one_shot=True)
  File "C:\Users\dmf\projects\invest\env\lib\json\encoder.py", line 257, in iterencode
    return _iterencode(o, 0)
  File "C:\Users\dmf\projects\invest\env\lib\json\encoder.py", line 179, in default
    raise TypeError(f'Object of type {o.__class__.__name__} '
TypeError: Object of type Unit is not JSON serializable

We actually have a test for this in test_model_specs.py and a special serializer in spec_utils.serialize_args_spec to format pint.Unit objects. So the problem is likely just that the CLI is not using our serializer, but just trying json.dumps instead.