If you already have two probes, just run with that. Just make sure your extra probe you've been storing was stored properly, else it's toast anyways. If it is toast, then just order the lab grade probe and use it on the reactor.
You have the correct idea for controlling the CaRX. I tend to keep a tighter window than 6.4-6.8 for the solenoid. I currently am at 6.5-6.6 and my solenoid only turns on/off every 3 hrs. That's actually amazing as I usually have it actuating 2x/hr. You want to not have to actuate too much and try to have the setting on the regulator close enough to not even needing the solenoid to turn on/off much... that way if your solenoid ever stops working, which mine has done 2x in 4 years (so it happens), your system won't be flooded with CO2. Hopefully your controller alarms will go off to warn you of that but it's a nice failsafe to have the bubble rate close to what it's supposed to be anyways and not have the solenoid do all the work for you.
I have two regular probes. If one ever doesn't calibrate well anymore or drifts too quickly after calibration, then my replacement will be a lab grade version. Either way, I only calibrate 2x/year and it doesn't drift too bad so Jim's experience with his regular probe might be more the exception than the rule.
For your question on precision of the reactor ph probe, as precise as you can get. Even when I make a simple change of 0.02 pH on my controller, it has drastic effects on the amount of alk/Ca I am putting into my system through the reactor. The more precise, the better in my opinion. Hope that helps!