[x] create toggleable legend with 2 variables in check boxes
[ ] create a bivariate color ramp when both are selected
[x] experiment with data options - peak swe, snow or peak swe date with the second variable being either the diff or anomaly from the historical record
From these explorations, I have concluded that the bivariate legend may be overcomplicated for the message. Pivoting to make 2 data layers that are anomalies for magnitude and timing.
id
similar to temp prediction (https://github.com/usgs-makerspace/temperature-prediction/blob/master/src/views/set2/Set2.vue#L1011-L1037)