investigate performance regression

[myoung3]

intervalaverage v 0.8 and data.table 1.12.8 is faster than v 0.9 and 1.13.x

when running a large intervalaverage (mesa events analysis) . this might be due to the addition of maxgaps, or possibly data.table performance regression.

create a publicly shareable reprex and identify what's causing it.

[myoung3]

My current thinking on this is to revert maxgaps for now since the project we were planning on using it for may end up relying on postgres procedures instead. maxgaps will still remain on the feature list, but cleaning up the C code and i particular adding better precision for sums is higher priority.