return_ci=True is not properly passed in get_ate_ci of the Sensitivity class in sensitivity.py

[lee-junseok]

Describe the bug
Some learner's estimate_ate function requires return_ci = True to return ate_lower, ate_upper together with ate. get_ate_ci of the Sensitivity class does not pass return_ci = True to estimate_ate.

To Reproduce
Steps to reproduce the behavior: Run sensitivity_analysis method with a class object Sensitivity(..., learner = XGBTRegressor(...), ...)

It gives
ValueError: not enough values to unpack (expected 3, got 1)

Expected behavior
Get sensitivity analysis result without a ValueError

Screenshots

Learner: XGBTRegressor()

Environment (please complete the following information):

• OS: [Ubuntu (Jupyter notebook)]
• Python Version: [3.7]
• Versions of Major Dependencies (pandas, scikit-learn, cython): [pandas==1.35, scikit-learn==1.0, cython==0.29.34]

Additional context
None