Well, complex issues like this take a lot of differential testing to disentangle, and my efforts are modest.
I did set up the full-spectrum Z6 with UV-Nikkor 1054mm f/4.5 lens, using the Baader U(2nd.gen.) UV passband filter.
The first test shot was done in April this year outdoors. Striping was quite evident at that time (100% crop for ISO 100 shot below). This camera had the original v.1 firmware. I updated it yesterday to v.2.2. Of course no flowers can be found outdoors in December here when snow covers the ground, thus a section of my book shelf had to serve as test target. Entire frame and 100% crop shown below.
As stated in my earlier post, there still is some striping phenomena in the output when one shoots UV. Visible-light only, or IR, seems to fare much better and apparently one would not likely encounter striping issues there (to be better tested). I tested various RAW conversion programs and Capture NX-D, Photo Ninja (latest version 1.38b), and Silkypix 8; all did exhibit faint traces of the striping, but not severely so. All of these programs did a good UV "white" balance. RawTherapee 5.7 can remove all signs of striping using its dedicated built-in filters, but cannot handle UV "white" balance properly. A classic "between a rock and a hard place" scenario.
image examples here are with Silkypix. Shot at 100ISO, using a Broncolor studio flash with uncoated Xenon tube as illumination source. First a 100% crop from v.1 firmware, then v.2.2.