introduce `ltry`

[bcjaeger]

currently oblique forests just try to make one linear combination of predictors, and they will try again (with different predictors) if that split isn't good enough.

There should be more flexibility with this. What if I want to try 5 different linear combos and pick the best one?

This would change some of the core C++ routines in splitting data.

[bcjaeger]

the API change should be straightforward. Include the input ltry in orsf() with a default value of 1.

So orsf(ltry=1) would do what orsf() has always done, but orsf(ltry = 3) would pick the best combo of the three combos assessed