This is a fascinating and elegant little virtual machine design that I think many here would find interesting.
Altough the complexity would skyrocket.
Then, if I wanted to write an adventure in Spanish I would omit the tildes, starting question/interrogation symbols and I would map ñ to ny.
That's not an inherent limitation of the VM at least, just of the current software ecosystem. It would be possible and practical to make a tiny unifont-based Unicode text library. Though handling bidirectional text, joined scripts and vertical scripts would be something else… :(
Some fonts for uxn (look at the ~rabbit repos at git.sr.ht) already bring extended chars.
Making a bespoke ASCII extension would be a step back - by about 30 years. You don't need a lot of code to support UTF8; if you're concerned about runtime memory usage, you can make your rune type take 8 bits and support only the U+0000-00FF range[1]. It happens to cover all of [áéíóüñ] and a whole bunch of other languages - unfortunately, not my native one, which would leave me gravely upset ;P
Unicode is also just a representation for text, and a handful of common operations - you work with arrays of characters, rather than arrays of bytes. It was worth its cost on 1992 hardware; Nintendo DS is over a decade more recent.
I recommend studying libutf in sbase[0]. It's not a single header file solution (although utf.h[1] is an excellent place to start reading), but it does provide a fairly comprehensive implementation. There's also a good introduction to Unicode in Plan 9's C programming guide[2]. Even if you choose to only support runes that fit in a single byte, you gain the ability to tell byte blobs apart from text, which is useful both for reasoning about your program, and for future-proofing it, in case you needed to put places like Łódź or Πάτρα on your map.
[0]: http://git.suckless.org/sbase
2. Sometimes it might, but that has nothing to do with uxn. Sometimes it isn't helpful to be differentiated anyways, and sometimes this differentiating byte arrays from text causes problems, too (it isn't really so uncommon).
3. The niju program does not use Unicode and does not need it; it works better without it. If you do want more sophisticated Japanese text, even then there are better ways than using Unicode.
I design my own programs and specifications to avoid Unicode as much as possible, even when multilingual text (sometimes even in languages that Unicode does not have) is desirable.