I have several of those amps, which I have not yet put into service, and I am planning to drive them with either an IC-705, which probably maxes out at around 10W out on 6m, or else I also intend to drive them with a recent-vintage Hermes board from Apache Labs, driving either the companion 10W amplifier (which may possibly have reduced output on 6m), that is, if Apache Labs ever gets around to fulfilling my order that I placed a long, long time ago (for which they are not responding to my customer support inquiries, which is par for the course for Apache Labs), or else I will be driving a 10W amplifier kit that I got from QRP Labs (which I have not built yet).
Neither of those options should encounter the transmit spike problem.
However, depending on the output power that I see from driving the Harris with either of those radio/amplifier combinations, and depending on whether I may decide to drive a pair of the Harris amplifiers (with a splitter on the inputs and a combiner on the outputs), I may later decide to use my IC-7300 as the driving rig.
Either way, I will be using an older W6PQL sequencer to handle the sequencing.
So take my comments with a grain of salt.
But if it were me, I would not want to hot-switch the input to the Harris via the sequencer as a way of protecting against a possible rogue spike.
So, in light of the potential severe consequences should such a spike become an issue, personally I'd be more comfortable running the IC-7300 at full output, and putting an appropriate attenuator between the output of the IC-7300 and the input of the Harris.
Yes, the attenuator will get hot - but either a fan or a higher power rating for the attenuator should take care of that. You might have to experiment with different attenuators to get things 'dialed in' to your liking.
I'm hoping (and hoping) that I'll be ready before the June VHF contest. But that was my plan last year as well