fixes errors when computing ME's for factor features

[scholbeck]

• created two functions for modifying data and predicting on new data: one for factor features and one for numeric features
• computeAME() now only accepts numeric and factor features -> check when using function
• added unit test for this new feature (data type check)
• removed derivative functions for character and logical features
• the calculated derivatives are highly sensitive to the chosen prediction function -> user gets notified in updated documentation
• adjusted unittests such that mlr, caret and glm function return the same AME's (depends on the supplied prediction function)
• adjusted margex.Rdata file, such that it only contains numeric and factor features (previously included an integer feature)

--> this fixes both errors mentioned in issue #12

[codecov[bot]]

Codecov Report

:exclamation: No coverage uploaded for pull request base (master@e48eb29). Click here to learn what that means. The diff coverage is 100%.

@@            Coverage Diff            @@
##             master      #13   +/-   ##
=========================================
  Coverage          ?   88.11%           
=========================================
  Files             ?        4           
  Lines             ?      101           
  Branches          ?        0           
=========================================
  Hits              ?       89           
  Misses            ?       12           
  Partials          ?        0

Impacted Files	Coverage Δ
R/computeAMEInternal.R	67.64% <ø> (ø)
R/derivative.R	96.66% <100%> (ø)
R/computeAME.R	100% <100%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update e48eb29...907f2ca. Read the comment docs.