Support for one-sided hypothesis tests

[nicolaiberk]

Hi,

thank you for providing this great package!

I am doing a lot of experimental work, and often I try to pre-register directional hypotheses in order to reduce the necessary sample size. I would like to model this using lm_robust and difference_in_means, but at the moment I can only work around it using the output from lm_robust in downstream calculations.

A feature implementing one-sided hypothesis tests in these functions would be amazing!