Improved input validation

[dbrakenhoff]

Users yesterday did a proper test of our input validation, and it didn't always lead to informative errors :). So there are some improvements we can make:

• topboundary="phreatic" in a steady model is not supported, but gives a weird error when you put it in
• Creating a model with a single z-value doesn't make much sense, but also leads to a cryptic error.
• Setting parallel=True but no tqdm installed silently falls back to single core computations
• ...

[MattBrst]

And the ngvbc issue that we had?

[AlyanPremani05]

Hello @dbrakenhoff @MattBrst ! Would love to work on this issue. Could you please assign it to me.

[dbrakenhoff]

Sorry @AlyanPremani05, I appreciate the offer, but I want to give the opportunity to one of the users first.

[dbrakenhoff]

And the ngvbc issue that we had?

Ah yes. model not solved/initialized, that will be a separate issue.

[BlackPool25]

Hii @dbrakenhoff @MattBrst !
Could you please assign me this issue?

[BlackPool25]

Taking a look at this. Planning to add explicit validation for the three cases: validate topboundary values, check z array length, and make the parallel-fallback warning visible.

[dbrakenhoff]

See my answer above @BlackPool25.

[dbrakenhoff]

Note to self: do not label issues with "good first issue"...