3D printing. Of woven cloth BDUs. In the field.
Yeah, no. The problem isn't how or where you manufacture the BDUs. The problem is suiting them to a specific environment. Active camouflage can theoretically change on the fly to suit any location at all, but passive camouflage has to be suited to work in various different locations within a general region.
Soldiers move a lot. They cross a lot of terrain, and they need camouflage that operates just as well in one spot as it does 100 yards away.
A tree bark pattern works a lot better than other patterns when you're standing against trees (of the correct species, of course), but it doesn't work so well when you're in dense underbrush, or on top of lichen covered rocks. If you're slogging through the dense woods somewhere, you're going to encounter all three sub-conditions of the woodland environment in close proximity to one another, and you need camouflage that works in all of them.
Militaries already extensively research specific environments and then tailor their patterns to work throughout those regions. Woodland camouflage is meant to work reasonably well across all the various sub-locations you might find within a woodland area. It wouldn't make sense to produce more specific patterns than they already do, because then you run into the problem of the tree bark pattern while in underbrush, as mentioned above.
Active camouflage can overcome this limitation because it can change as often as you need it to, and adjust to changes as small as a few inches.
There's just no reason to do as you suggest. 3D printed cloth is almost certainly not going to be of a sufficient quality to serve as BDUs, and is definitely going to be more expensive than traditional textile loom weaves for many years to come.
There's no reason to manufacture the garments in the field either - even if you wanted to change passive camouflage frequently, you already need to have a supply line, just add the alternate BDUs to that logistical system.