clarify NoiseAnalysis

[kainagel]

The class NoiseAnalysis, which is used from the noise analysis branch of the application contrib, probably does not exclude bicycles. This needs to be understood and clarified, and then the most typical use cases either need to be included by default, or made configurable.

It seems that all modes that are not explicitly excluded are classified as "pkw". I am wondering a bit if we should maybe proceed as in emissions, where the emissions vehicle type needs to be in the vehicle attributes.

This should then possibly be set in the scenario runscript (so that allVehicles already contains the correct information). (The same would then hold for particulate emissions.)

Related:
#3520

To check:
#3328 #3315 #3495

[tschlenther]

From talking to @GregorRyb, I understood that

1. part of the problem was the issue that every vehicle was assumed to be a car, if not explicitly declared otherwise. This issue was addressed in PR Don't assume car network mode as default #3532, which removed the default value ('car') for each the networkMode in VehicleType. .
2. the PR typical vehicle categories for noise analysis #3520 introduced default settings for noiseParameters.setNetworkModesToIgnoreSet and noiseParameters.setHgvIdPrefixesArray when run from NoiseAnalysis.

I now added #3557, which

1. sets a default for ignoredNetworkModes directly in the NoiseConfigGroup to the same set that typical vehicle categories for noise analysis #3520 introduced for the NoiseAnalysis. This means that the same default is used for any other setup, for example for online analysis or when another noise script is run.
2. clarifies comments in NoiseAnalysis on the default settings.
3. introduces consistency checks on ignoredNetworkModes and hgvIdPrefixes, including warnings when the defaults are not in use and an abortion in case the defaults are not in use AND one of the walk or bike modes is indeed a network mode.

I hope and think that this issue can be closed as a consequence!? Let's discuss :)