Decker is open-source, so nothing is impossible if someone is sufficiently motivated to tinker with wild ideas in their own fork.
That said, I don't think porting Decker to the 3DS would be at all easy. Decker's user interface is designed with a minimum resolution of 512x342 pixels, while the 3DS screens are functionally 400x240 and 320x240 pixels. Decker is designed to be usable with a stylus and no extra buttons, but only the lower screen of a 3DS can respond to touch, and you'd probably need that entire touchscreen to serve as a keyboard whenever you want to edit a script, rename a widget, or generally interact with a field. The 3DS has very little RAM and a very slow CPU compared to most of the computers people typically use Decker on, so you'd need to do some careful optimization and possibly remove or otherwise constrain some features entirely, like the unlimited undo history or the ability to have hundreds of cards in a deck.
Performance aside, I think you'd need to completely overhaul Decker's UI to suit such a device, and it would be difficult to run the decks and interactive docs people have already made for Decker's normal resolution (or bigger). It might be more technically feasible to make some kind of a stripped-down "Decker-player" for handheld consoles that exclusively runs premade decks designed with a lower screen resolution and lighter specs in general in mind, but it would be kind of a shame not to have the editing tools or the existing library of decks...