adding code for existing scikit learn model

[egillax]

• code that lets users plug in scikit learn models into the PLP framework

[codecov]

Codecov Report

Attention: Patch coverage is 92.23301% with 8 lines in your changes missing coverage. Please review.

Project coverage is 87.54%. Comparing base (a6e87de) to head (a5b8994).
Report is 2 commits behind head on develop.

Files with missing lines	Patch %	Lines
R/ParamChecks.R	84.44%	7 Missing ⚠️
R/HelperFunctions.R	85.71%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #531      +/-   ##
===========================================
+ Coverage    87.44%   87.54%   +0.09%     
===========================================
  Files           49       50       +1     
  Lines        10913    10967      +54     
===========================================
+ Hits          9543     9601      +58     
+ Misses        1370     1366       -4     

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

[egillax]

@jreps I've made a PR from your branch with the code to convert a sklearn model to PLP.

• I simplified the covariateMapping, AFAIK we don't need any sort of covariateNames associated with the sklearn model. In sklearn there's no info about the features saved with the model. We just need to map from our OHDSI covariateIds to the correct column the sklearn model expects. I added a test using reticulate and sklearn to train a model and test that predictions are identical between PLP and sklearn:
  

  PatientLevelPrediction/tests/testthat/test-existingModel.R

  Line 146 in a5b8994

  expect_equal(mean(prediction$value), mean(externalPredictions))

• Added quite thorough tests and checks.

• I simplified naming - got rid of redundancy and added more explicit shorter names where possible (IMO).

• Some minor unrelated changes with regards to helpers and parameter checks.

Please review and let me know what you think!

[jreps]

Looks good to me! Nice idea with removing the names for simplicity.