I don't know how many have GT front brakes, but I'll ask anyway.

I have the disk height measurement (from the mounting surface between the top and hub to the back or rear surface that the pad presses against) at 38 mm. Correct rotor thickness for new is apparently 10 mm.

I am using new wheel bearings, inner is PN 1988 with a PN 1922 outer race (which is what came off the car).

I have bolted the rotors tot he hubs, placed the hubs on the spindles (and tightened the axle nut), and bolted the caliper cage onto the spindle.
I am measuring the space between the caliper body and the rotor. I am ignoring the pad guides, which are adjustable, and measuring from the slot machined into the caliper body. That slot is just under 14 mm wide, which would leave 2 mm on each side under ideal conditions, and this seems like a remarkably tight fit on a 40 year old car.
I measure the caliper mount surface to the rear rotor surface to be 18.56 mm.

For the rotor to caliper body clearance:
I come up with a measurement of 2.286 mm on the outer side (toward the wheel).
And 1.803 mm on the inner side (toward the chassis).

So it is about .2 mm toward the inside. Which looks like a huge difference in such a tight space. If it were a long sleeve shirt, it would be like having the left sleeve end at the elbow and the right sleeve end at finger tip length.
My concern is if this be a change in the bearing thickness or a normal manufacturing variance.

Anyone else notice this?

i'd say 99.9% of us that have GT's have GT brakes JT!
that variance is normal. u should see the difference on some cars....
i know this sounds silly, but everything has gone back together squarely? and u are using new bearings, and i would think, bearings cups.... have the cups been competely seated in the hub?
honestly tho, it equates to 0.1mm of wear tolerance in each of your new bearings, which as both know, do wear over time, meaning that as the bearings are re-adjusted, the hub will move (very slightly, by fractions of a mm) towards the centre of the car.
this then makes the spacing exactly central, as starting with a perfectly central spacing will make an offset to the inner edge over time if set that way.
i personally wouldnt be concerned over 0.2mm overall, especially as ive seen stub axles (what u call a spindle) bent more than that due to 40 years of use, and they then have more difference than 0.2mm.
cheers.

Well, there is a much larger number of sedans and coupes than GTs... And a lot of people switched to different brakes...

To get the clearance tighter on the back side of the rotor (like what I am finding with my measurements), I would have to over press the inner race into the hub. I've mashed a few things with the press in my time, but I hope I learned and I was trying to be as careful as I could. I don't think the design would allow it to be over pressed without distorting.

If there is any wear on the axle shaft where the inner bearing seats, it is worn so uniform that it looks as if it was made that way.

I started in on this and asked myself if I wanted to correct any manufacturing errors back to optimum or if I wanted to correct it back to the way it was originally. If anything that I do is to be interchangeable with another car, it has to be back to original, or my correction to optimum on my car might be a correction outside of what would fit on another car.
That's one of the reasons I like to pick up a spare or two from the salvage yard or work from measurements from a couple different cars over the life of the production run...

Something else I stumbled over was the difference between the specified "nominal thickness" and the actual measured thickness. The inexpensive, new Nissan 280ZX rotor I bought (accurate size, questionable alloy), has a nominal thickness of 20 mm. It has an actual thickness of 20.2 mm. And the extra thickness is on the outward facing side.
If the Bellett GT rotor nominal thickness is 10 mm, and the actual measurement is 10.2 mm, then it would be perfectly centered in the caliper bracket.