upper CI is truncated to 2 decimal points, lower and point estimates aren't

[tingleyd]

This has caused some problems from some users.

Can be replicated from the package example file (below). I suggest using the same rounding for upper as we do for lower.

summary(contcont)

Causal Mediation Analysis

Quasi-Bayesian Confidence Intervals

           Estimate 95% CI Lower 95% CI Upper p-value  

ACME -0.0183 -0.0397 0.00 0.04 *
ADE -0.0450 -0.1417 0.03 0.40
Total Effect -0.0634 -0.1636 0.01 0.16
Prop. Mediated 0.2686 -0.9015 2.54 0.20