C-API: crg_check function unused

[CarawaySeed42]

From 3dbastian:

The current state is, that crg_check is implemented in C but not used.

Invalid CRG-files can thus be read and evaluated.
Question: Do we want to implement the crg_check on read or do we leave the decision to the user/caller?

[jorauh]

Since the implementation of crg_check (and interpreting warnings and errrors detected) is quite complex, the former OpenCRG project decided to leave the check to the matlab toolset.

So we should advise the users, that as long as a crg file is not pre-checked by the crg_read of the matlab-toolset, its use is risky and may fail in the c-api without useful warnings or error messages.

Since crg files (especially the clear text header) can be manipulated by a simple editor, it does not thelp to implement some trivial feature in the c-api looking for any message of a previous successful check.

[CarawaySeed42]

For reference, as of now, the crgCheck function in the C-API checks the crg options, modifiers, the curvature, if the refline can be closed and if the dataset exists. Information about encountered issues are just printed to std or using a custom print function.
The C-API check is very incomplete compared to the Matlab-API check.

Communicating this and that there are no critical data checks to users might be helpful. Communicating as in documenting.

A forced check during/after load is not very helpful. Some of the existing check results are just nice to know and don't lead to critical errors.

So, I am of similar opinion as @jorauh