testTwoRFs: Test two random forest classifiers
In kkatellis/TLBC: Two-level Behavior Classification

Function to apply two random forest classifiers to some data.

1	testTwoRFs(featDirs1, featDirs2, rf1, rf2, saveDir, testNames)

`featDirs1`	Path to a directory (or list of directories) containing features corresponding to the first RF classifier.
`featDirs2`	Path to a directory (or list of directories) containing features corresponding to the second RF classifier.
`rf1`	A random forest model.
`rf2`	A second random forest model.
`saveDir`	Path to a directory where predictions will be saved. Predictions will be saved in files named <identifier>.csv with two fields: timestamp, prediction.
`testNames`	List of participant identifiers to use.