rT.variables in mg/dL

[mountrcg]

First part of fix for #29

Check in Trio needed to find where suggestion.isf is used and apply a glucoseformatter to it. I did not find deviation and BGI being transferred to Trio from oref.

[mountrcg]

Checking on rT.target_bg, there seems to be an error in the calculation of insulinForManualBolus

trio-oref/lib/determine-basal/determine-basal.js

Line 1623 in d1dfb70

rT.insulinForManualBolus = round((rT.eventualBG - rT.target_bg) / sens, 2);

as it uses a non converted glucose value eventualBGand a converted glucose value target_bg, which will screw up that value for mmol users. Changing rT.target_bg also into non converted value - it duplicates current_bg though.

[mountrcg]

successful test documented in Trio PR nightscout/Trio#327