Support Selective OneHotEncoder

[thismlguy]

Currently the OHE encodes all the values, but there might be 1000s of them which will blow up the training data. It'll be good to have a control to limit the number of columns generated after OHE.

Some solutions from the discussions on slack:

1. Have a parameter "N" which defines the number of columns and most occurring N categories will be kept.
2. Have the list of required categories be defined in the feturespec. The user will define a list of N categories required and only those will be OHE, others ignored.
3. Have a percentile defined, e.g. keep all categories accounting for say 90% of the observations.

From initial discussions it seemed, #1 and #3 are harder to implement that #2? Open to discussion.

[richwhitjr]

1 can also be done without too much effort using some of the heavy hitter monoids we use in other transformers. 3 can be difficult since we don't know how large to make the sketchmap in the first place to pick up all of the counts to calculate the distribution.