Difference in p-value precision in t-tests

[emil-hagglund]

Hi,
Thanks for a nice package. I have observed difference in the precision of p-values reported depending on the number of groups you have when you perform t-tests. I noticed that pairwise_two_sample_test() uses p_round(digits = 3) before returning the results, but this is not done for two_sample_test(). Is there any specific reason for this? I get different results if I want to adjust the p-values downstream of performing the t-test and would prefer if the p-values where not rounded before they are returned.