Cyanotypes at first reminded me of the prussian blue marking liquid I used to use in shop to mark where I'd be cutting steel. It would make scibe marks show up well, and stuck real well to the metal. It's apparently some sort of dye mixed with laquer, and it stuck to steel super well. I haven't tried it with aluminum. I bet cyanotype chemical could be made to work somehow, just prevent the liquid from beading up as it dries.
I like Pauls's resist+anodizing idea! You could do multiple steps like silk screen! I used to use mylar and a laser printer for making resist masks for copper circuit board etching. I ironed it right off the mylar and onto the metal.