resultsExplore and test

[cebarboza]

Here is the first attempt for resultsExplore. I added a test in test-MainFunctions. As you mentioned, it is a good idea to access any result without running everything again. I left the option to get several results in testExplore, so it is easier to not call resultsExplore in every test.

[AniekMarkus]

Great that you have added resultsExplore! I do think it would be useful to have as output of that function this list format (can be extended with more items, but I think these are most important):

results <- list("model" = rule_string,
                  "candidateModels" = candidate_models,
                  "countCombinations" = total_count_combinations,
                  "countFeatureOperatorPairs" = total_count_feature_operator_pairs ,
                  "countRulesWithoutConstraints" = total_count_cutoff_sets,
                  "countRulesWithConstraints" = total_count_candidates_(incl_constraints)
)

For the trainExplore function I think we should choose the standard output to be what users expect to be returned; i.e. the model so have resultType = c("model") as default? Then you can still use the other resultTypes for testing purposes. That way it also still nicely works when calling predictExplore afterwards.

@cebarboza What do you think?