Handle non-linear gain/ADC overflow

[ckierans]

The gain in the ASIC becomes non-linear before the max ADC/overflow bin is reached. This is not calibrated, so the easiest thing will be to implement a cut in ADC ~14000 (estimated from Clio and Brent May 2025) and flag these hits as overflow.

[ckierans]

This was brought up again at a calibration meeting Jan 30, 2026 when talking about recovering high energy photopeaks.

Specifically: We have to flag these hits in the energy calibration. And we need to know where the nonlinearity starts for each channel - does this get calibrated from NRL for each channel? Then Julian has to handle those in the strip pairing.