This function calculates a multi-metric index based on the data
in SoH dataframe passed in. A tibble is returned containing a state of
health index for each day. The returned tibble contains columns:
datetime, index, and index_bin.
The index column contains a value normalized between
0 and 1 where 0 represents low confidence in the sensor data and 1 represents
high confidence. The index_bin is one of 1, 2, or 3 and represents
poor, fair, and good data respectively.
The index is calculated in the following manner:
If the A or B channel percent reporting is < minPctReporting, index = 0
Otherwise, index = pm25_A_pm25_B_rsquared
The breaks are used to convert index into the indenx_bin
poor-fair-good values.