Remove ringdown

Bayestar requires that the SNR series epoch matches the end_time of the trigger/IFO (I haven't checked which)... AND Bayestar requires that the SNR peak is in the center.

There's a good chance I've misunderstood the requirements above, but considering that our end_time has always meant the merger time, I'm guessing the original (postcoh) code was correct. SNR series epoch should correspond to end_time rather than snr_peak_time.

I'll confirm and can update if not.

For now, let's just get bayestar skymaps working. Even if this version is the 'wrong' one, it will affect all snr streams equally, and only by a fraction of a second, so the effect should be insignificant.