Convert multimeter test from SLI to Python implementation

[heplesser]

This PR provides a polished version of the Python implementation of the multimeter test ported from SLI.

[heplesser]

I am not sure why the Python tests fail here. It is again the numpy datatype problem, it seems, but that was fixed with #2571 and this PR does not touch that code. Also, all tests pass on my branch (https://github.com/heplesser/nest-simulator/actions/runs/4505551444) and locally on my machine.

[nicolossus]

@heplesser To me it looks like it is the static code check that fails? See comments. I also made a suggestion for how we can obtain models with certain properties.

[heplesser]

@nicolossus The static checks now pass, we now install an up-to-date pycodestyle via pip; previously, we took an old version from a system location, thus the trouble with the :=.

I don't think the models() function an improvement in this case so I have not adopted it yet.

[nicolossus]

@heplesser I see, thanks for adding an up-to-date installation.

I agree that the models() function (with a better function name) would fit better in a helper library. However, I still think that the if-statement

if ('recordables' in (dflts := nest.GetDefaults(model)) and dflts['recordables'])

should be changed to (in my opinion) the more concise and readable

if nest.GetDefaults(model).get('recordables')

(Note dict.get(key) will return None if key is not in dict, which is in contrast to dict[key] that will raise KeyError)

[heplesser]

@heplesser I see, thanks for adding an up-to-date installation.

I agree that the models() function (with a better function name) would fit better in a helper library. However, I still think that the if-statement ...

You are right, I didn't realize get() could be used like that. Changed it now, much more readable indeed.

[nicolossus]

Thanks! LGTM.