Incorporate different types of meta-learners

[jialul]

Create an option of X-learner, T-learner, and R-learner for lasso and BART and allow user to choose different meta-learner types.

[xiaolong-y]

To do so we can look into the following two packages:

• X-learner (causalToolbox)
• R-learner (rlearner which also implements S-, T-, U-, and X-learners)

rlearner might be a good place to start with since it is an encapsulating package.

Afterwhich, we can also expand our support causality specific packages like doubleML. Some of listed here.

[jialul]

Added R-T-X-U learner in commit [571bb2e](https://github.com/MichaelLLi/evalITR/commit/571bb2e7bba2d2a15e742f4e735339ca88cfa2ae)