Vacuum is affected by rpm more rpm=more vacuum.
Do you use an A/F ratio meter on all the engines you build?.
Yes vacuum does increase with RPM assuming the load remains constant and is low. Example: Hook up a vacuum gauge to your intake manifold. While riding down a level road in 2nd gear, gradually increase your rpm until your engine is going 3500 RPM. Hold that RPM. Your engine vacuum will of course be high. Probably somewhere around 23 in/Hg. so what you said is True. But now quickly open your throttle to 100% and see what happens to your vacuum. It will fall like a rock to probably 10 in/hg. Why? It was at the same RPM but your vacuum gauge read a way different vacuum readings due to the load on the engine when you nailed the throttle. Want to see an even more dramatic example. When you were at your constant 3500 RPM in 2nd gear, shift into 3rd and nail the throttle to 100%. Your vacuum reading will be close to 3 in/Hg and you only dropped your RPM by maybe 500 RPM.
Performing the 1st part of this example (holding throttle at 3500 and snap it to wide open) with the bike in neutral is meaningless since you did not apply any real load to the engine which is what makes the vacuum fall on its face. And unless the mass of the rollers on an inertia dyno are large enough to extend the pull times out where they should be, you won't see it happen there either. Hence the need for actual road testing or an eddy current dyno.
Another good example of how RPM does not track with vacuum is in how a power valves works in a performance carburetor (old school Holley). They open when they sense low vacuum which equals to more load which equals to needing more fuel or needing a lower A/F ratio. Of course vacuum will change with RPM but power valves don't open and close based on RPM. They open and close based on Load demands and that is recognized through vacuum levels.
I asked but you did not say if your sole means of spark advance is your VOES. I have seen some aftermarket setups that have Base timing along with some additional electronic advance plus VOES.
I take it your total advance is 28 degrees with a base of 10 so the VOES is 18..(?) Since you have the toggle point of VOES set at 6 in/Hg, that would mean at cruise speed under light loads you have 28 degrees of advance. That is well within the ballpark and is by no means on the edge. I can not see how that would cause the pipes to glow. Since you can increase RPM and past through the glow point, I would continue looking at the carb or a vacuum leak.
To answer your question about A/F ratio meters. I don't build engines but yes, I always use both an Innovate wide band A/F ratio meter and a Bear/Marquette dispersive Infra-red exhaust analyzer. The Innovate I mount to the bike and data record while riding it. The Infra-red gas analyzer is a shop roll around tool I use while stationary tuning. Without them, everything is just a guess. It is not too bad with carburetors because carbs feed both cylinders with somewhat of the same mixture or A/F ratio. On fuel injection I can set the A/F ratio on each cylinder independently so I probe each tail pipe before the cross-over pipe. This is what I do and this is how I tune my bike.
I probably can't help you very much because I don't understand HD carbs to any great degree. I thought I may be of some help if you felt it was a spark advance issue. It sounds like your on the right track.