CO2 holds moisture like Oxygen does. Nitrogen does not. I didn't think of using 2200 or less PSI nitrogen to feed a compressor that would take it to 3000+ but I agree, that would avoid the need for a regulator that would handle 6000 psi and also provide a dry source of "air". Air is mostly nitrogen anyway so the gun should be completely compatible. Could be a good alternative if you have a reasonable supplier of the nitrogen.
But I still like my option of just putting a relatively small additional dryer filled with color changing dessicant on my YH. Didn't cost a lot and works fine. I have to open the air tube of my P35s to change the regulator and I had to do it once to change O-rings and there was no sign of moisture in either gun. After thousands of pellets and hundreds of fills. There could still be moisture in my tank, I guess, but it seems like there would be some sign of moisture in the guns if there was. Tanks is still holding 300 bar fine.
Target forge also had a lab test air from a YH and from a CS2 and with a dessicant filter, both produced air without significant moisture but only with an extra filter filled with dessicant. It's his episode 18 where he discusses the results. He tested an after market coalescing filter on the CS2 and it did not work at all. But he thinks that was due to the low air flow and he said he would add it to the YH to reduce dessicant replacement. The CS3 apparently uses this type filter. Coalescing filters depend on having enough velocity in the air and when they do, they can be effective. He tested the stock CS2 filter (which he apparently had been using) and it was terrible. High levels of moisture in the air. He also found evidence of dieseling from his YH filled with hydraulic fluid (oxygen was low and CO was high). If you are interested in this topic I recommend viewing the video.