Compute ideal sample size based on parameters

[jnu]

A feature request came in for setting the sample size automatically to get statistically significant results. I like this idea, with the following reservations:

• I don't want to clutter the tool with more widgets that exist (in multiple places, freely, and easily accessible) elsewhere on the internet
• I don't want to add a simplified widget that would encourage people to do bad statistics / give a false confidence in how solid their eventual analysis will be.

I could be ok with suggesting a sample size with some context tips about how we derived that number, and some caveats about how to use it.