Improve CI and some code cleanup

[dbrakenhoff]

See #356.

[dbrakenhoff]

Test fail related to HHNK map server...

[bdestombe]

I am a bit uncomfortable with skipping the KNMI test, but I am also not fully understanding why it fails.
File not found. Should we introduce logic that solves for the missing data? Or instead of skipping the KNMI test, should we alter the test as the test problem as it is posed now should fail? Or is the user never going to end up in a situation that this error would be relevant?

[dbrakenhoff]

The test skip is meant to be temporary. When @martinvonk is back from his holiday, I was planning to let him look into it. Meanwhile we can release the next version of nlmod without being bothered by these failing tests.

I thought the KNMI Data Platform functionality wasn't that extensively used, so it would be okay to fix later.

[OnnoEbbens]

The test that was skipped is for the KNMI data platform to obtain temporally and spatially distributed meteodata. In the examples we don't use the data from the KNMI data platform. We use the KNMI data obtained via hydropandas. If I remember correctly it is not very useful to use the data from the KNMI data platform yet as there are some weird interpolation results which might mess up the model. Also if you use temporally and spatially distributed meteodata in the model it usually becomes very slow. So we don't have a very good use case for this data yet.

[bdestombe]

Ah, I wasn't aware that this was an alternative route for obtaining KNMI data. Thanks both of you for clearing this up.