Fix spurious bumps on load current before RmpWfm start

[gabrielbrunheira]

During tests of the feed-forward rejection system for Septa disturbance, some bumps in the load current were observed between the arrival of a sync pulse and the start of execution of a waveform reference on RpmWfm operation mode, as captured with the osciloscope persistance and seen in the image below:

This behavior resembles a impulse response observed during tests of NLK disturbance on some FBP power supplies, and may indicate that a spurious not null single-point is injected in the controller loop (maybe in the the reference input). This hypotheses is reinforced by the repeatibility of shape and time instant that it occurs, as also observed in the image above.