Hotfix: assign centroids in cost benefit impact calculation if in new exposure no centroids were assigned

[ValentinGebhart]

Changes proposed in this PR:

• If a Measure included a new exposure object with no hazard centroids assigned, this led to an error message that was not clear how to resolve, see CostBenefit measures with new exposure need assigned centroids #1056. Now the method checks if there are assigned centroids and assigns them if not.

This PR fixes #1056

PR Author Checklist

• Read the Contribution Guide
• Correct target branch selected (if unsure, select develop)
• Descriptive pull request title added
• Source branch up-to-date with target branch
• Documentation updated
• Tests updated
• Tests passing
• No new linter issues
• Changelog updated

PR Reviewer Checklist

• Read the Contribution Guide
• CLIMADA Reviewer Checklist passed
• Tests passing
• No new linter issues

[ValentinGebhart]

@spjuhel FYI; but I guess not relevant for the cost benefit restructuring

[peanutfun]

Thank you for the fix!

I would appreciate a test. Could you just add the minimal example from the issue as test case? The test should make sure that the computation runs even when no centroids are assigned, and that the warning is issued. Alternatively, it might also be easy to add such checks to the existing unit tests for Measures and CostBenefit.

[ValentinGebhart]

Thanks, Lukas! I've added a small test that check that the calculation runs if no centroids are assigned and that a warning is raised, and that no warning is raised if centroids are already assigned.

[ValentinGebhart]

@peanutfun I have know changed the case when we check that the centroid warning should NOT be raised, as we discussed. I checked that it works (ie raises an error if the centroid warning is raised wrongly, and passes if no warning is raised). It somehow looks a bit awkward to me, but if you think this is good enough, we can merge.