Run cm analyses tests

[mvankessel-EMC]

I've added dedicated unit tests for runCmAnalyses().

I've added the following files:

1. setup-runCmAnalyses.R (mostly copied from test-eunomia.R)
2. test-runCmAnalyses.R

I've also changed the following in the existing code:

1. Replaced the for-loop checks for cmAnalysisList and targetComparatorOutcomesList to assert class "cmAnalysis" and "targetComparatorOutcomes". Both assertions now expect the list to contain those object types instead.
2. Added a warning message when analysesToExclude has 0 rows.

[codecov]

Codecov Report

Patch coverage has no change and project coverage change: -0.29% ⚠️

Comparison is base (e1b07c6) 88.31% compared to head (158bc82) 88.02%.
Report is 7 commits behind head on develop.

❗ Current head 158bc82 differs from pull request most recent head a7779c3. Consider uploading reports for the commit a7779c3 to get more accurate results

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #148      +/-   ##
===========================================
- Coverage    88.31%   88.02%   -0.29%     
===========================================
  Files           23       23              
  Lines         5415     5380      -35     
===========================================
- Hits          4782     4736      -46     
- Misses         633      644      +11     

see 3 files with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[schuemie]

Again great work!

I'm not so sure about the warning for a zero-row analysesToExclude. It seems just another valid way to say no analyses should be excluded. What is your reasoning for throwing a warning?

[mvankessel-EMC]

Again great work!

I'm not so sure about the warning for a zero-row analysesToExclude. It seems just another valid way to say no analyses should be excluded. What is your reasoning for throwing a warning?

My thinking was that if you do not want to exclude any analyses most people will pass NULL to analysesToExclude, or leave out the option all together in their function call. I think passing an empty data.frame instead is odd, and maybe (probably?) not intentional. Especially when the data.frame is created programmatically.

It could be a message instead of a warning, but I think informing the user on the behavior like this is useful.

[schuemie]

Ok, let's keep as is.