I've been using ATF so far for the leadscrews and slides.
It doesn't seem to want to stay on the leadscrew well, esp not above 40ipm.

I was having a look at he ATF that puddled off the Z-axis leadscrew onto the y-axis way cover. I was alarmed to realize its cloudy look was actually a metallic "sheen". I took one of my super-strong N48 magnets and found the fluid was attracted, in fact it dripped upward like 1/8" at it. Fresh fluid did not do this.

OK maybe it was from "the work", although there's not been much I've done with ferrous metals which would be attracted by a magnet. I cleaned the screw, ran it to the bottom, poured a load of ATF from the top all down the Z-axis leadscrew, ran the headstock up slowly, cleaned the puddle off the way cover. Repeat.

Still has metal flakes in the fresh fluid. Wondered if I totally overtightented the leadnut. Disassembled, backed off the 2 side screws on the leadnut like 3/16 of a turn, tightened the center locking screw, repeated cleaning procedure. There is now a perceivable backlash- when I lift the headstock, it rocks up just a teeny click. At this point I have the motor off and will not be reattaching it for the rest of what I'm talking through here.

Again, forms metal flakes. So I've cleaned the leadscrew here, I'm adding fresh fluid, giving it a chance to "take" into the nut, rinse with more ATF, but I keep seeing this metal sheen and it's attracted by a magnet.

I'm beginning to think the ATF is a horrible choice for lube. But I cleaned it and tried some 30W chainsaw bar-and-chain oil, but again, it's shimmering with metal. Looks like not nearly as much as the ATF but that's hard to compare subjectively, the fluid color alone can throw that off, and the amount of fluid added and the number of runs before it dripped off was inconsistent.

Am I right to be alarmed, or is this somehow normal?

But the 30W stays on the screw MUCH better. I see that. It takes more speed to sling off, doesn't drip as fast, and I can see a thicker layer lingering on the leadscrew.