Timestamp column can be converted to type "object" which potentially causes the `.diff()` to fail

[samuelwnaylor]

wind-up/wind_up/scada_funcs.py

Line 31 in 980f6db

diffdf = df.groupby("TurbineName", observed=False).ffill().fillna(0).diff()

Example error when the .diff() is run: TypeError: unsupported operand type(s) for -: 'int' and 'Timestamp'