[LRNRQ] Add <regrssion_forest> from package <grf>

[chenx2018]

Algorithm

<regrssion_forest>

Package

Supported types

• classif
• clust
• dens
• regr
• surv

I have checked that this is not already implemented in

• [ x] mlr3
• [ x] mlr3learners
• [ x] mlr3extralearners
• [ x] Other core packages (e.g. mlr3proba, mlr3keras)

Why do I think this is a useful learner?

In casual machine learning, this learner is helpful to predict nuisance functions such as propensity score. If we have this learner, we can compare it with other common learners directly.

Further Optional Comments

(e.g. if this algorithm is already implemented under a different learner then please explain why this additional implementation is useful)