Small mistake in calculating P-value

[StevenZhangJW]

There are a small mistake in calculating the p-value from the delta methods, where
https://github.com/tunsmart/interactionR/blob/881b30b2048f8f97286840e1d26d471ea32c0300/R/interactionR.R#L370C3-L370C38

p.RERI should be 1 - pnorm(abs(RERI/se_RERI))
otherwise, when RERI is negative, the wrong P-value will be calculated.