Erik-- definitely try the coolest temp suggested by Ilford to see it helps with the contrast problem. Don't believe l ever once used P-30 at 92˚F, myself, having been alerted to the potential of increased contrast. So from me, it's just a well-circulated rumor, but it may indeed be the case and seems worth a try. Another side benefit of 75˚ is that I'd get more life out of the mixed chemistry, it'll oxidize slower at lower temps. Partly used bottles of bleach sitting in my processor with it running all the time were good for all of couple of days, tops, at 86˚.
I think I stand corrected on the process time at 86˚ though... just went and double checked the datasheet pinned on my D/R wall, and yep... 3 minutes, not 2.
The color balance of an individual paper pack should all be consistent, yes... Of course you have your paper pack filtration correction--which is usually a reasonably close approximation but not precise-- and your own enlarger filtration factor, which can change radically from what worked with a dying old bulb to the fresh new one you just replaced it with after dialing everything in. Welcome to the world of color printing.
Further, if you're are making big masked IC enlargements from 35mm, or for that matter any enlargement where exposure times run over a minute, reciprocity failure can start to cause minor shifts, so having the film technical data sheet at hand can be rather helpful, saving paper and bulbs on the really long long exposures. Uncorrected density/saturation shifts will be large beyond about one minute. This is from memory and might be a little faulty, but whole-stop progressions in reciprocity-failure-corrected time with CPM1M go something like 60 seconds to 180 seconds to 16 min to 45 min... filtration wasn't as weird or non-linear, fortunately. One key to consistent processing with Ilfochrome was standardize around one enlargement time as a base exposure, varying stops by aperture not by large time shifts, at least for the smaller common sizes. This all went out the window with bigger prints but 5x7's and 8x10s and unmasked 11x14s became relatively simple, very straightforward once I got a system down. It's reasonable to expect to make consistent prints. I was often making the same print 8-up and selling them side by side. They had to be really consistent.
Not being familiar with that Beseler color analyzer I can't help you too much there. I've only ever used an older Jobo ColorStar, which was an analog computer/ enlarging timer/simple match-needle --blinking LEDs, actually-- analyzer with adjustment sliders in only M&Y. Cyan was the base exposure level. Inscrutable as it was, it did prove useful as a metering timer, and for density measurements for estimating what strength masks I'd need.