Two words…
Swamp Cooler.
Get one of those evaporative coolers and set it up at the opening in the garage door blowing back into the garage. It might not make it chilly in there in the summer, but it should keep your machine from melting. If I remember correctly, they’re supposed to work quite well in a ‘dry heat’ type of situation.
You can also look into PC+ filament, it has a higher melt point and is more rigid than PETG but not as rigid as PLA. You have to have an all metal hotend and maybe a box over it to print (although with your ambient temperatures maybe not!)
We have a whole house evaporative cooler on the roof. In Denver, they work great. I have to do a little maintenance in the spring and fall, but compared to AC, it is energy free.
Went ahead and printed the new parts in PETG, installed them, and went for a night cut instead.
Its pretty nice to just “erase” the mistake layers and start again on the same piece of wood.
The stepper current does not have any influence on the heat that the (I assume part of the Z-holder) rails suffered from, visible in one of the last of the example images, though, me thinks.
Cool. Verbally.
Agreed. That is all heat from the ambient and the spindle.
The spindle vents blow directly on the z carriage, so yeah, that’s probably the culprit.
Steam spindle? 
For one of my 3D printer designs, I was having some heat deformation issues similar. Eventually I updated the design, but I had good success with “painting” the parts with polyester resin. A later run I used epoxy resin, because polyester resin is inherently nasty stuff, but I’ve got gallons of it from doing custom fibreglass stuff for car audio installs. A thin coat of resin was enough to make the part resist deformation. YMMV
What temperatures were you putting it at?
Currently, I can’t get below 100 degrees ambient during normal business hours ( I feel bad running the machine at night in my neighborhood) I can run it for about 30 minutes before the steppers get to 130. Then I stop.
Sounds like you live here, in Arizona…
