Remove statistics argument from add_stats?

[WillemSleegers]

I was thinking that the statistics argument does not make a lot of sense, and is potentially annoying.

The statistics argument can be used to only save some of the statistics, by providing a vector of the statistics you actually want to add.

Philosophical problem: This goes against saving all your statistical results. Pragmatic problem: If someone decides not to save, say, the t-value of a t-test, then the report function will not report it correctly.

Perhaps we can just remove it?

[arnoudplantinga]

I don't think I would ever use it (hard drive space is not really a problem), so I'd be fine with that!

[WillemSleegers]

Removed it.