I'll have to track down the root of the bug, but there are several best practices that you could follow that would prevent you from running into it at all.
You are using the 'Shown' Set with VSPAERO. While this is very quick, convenient, and easy -- it is not the way to go for any serious work.
Instead, you should establish a Set to use for thin surface components and another for thick surface components.
Next, there is no reason for the items inside the aircraft to be in the Set passed to VSPAERO. I.e. the pilot, fuel cells, systems, etc.
The root of the bug appears to be OpenVSP processing the custom components (fuel and systems) into thin-surface representations. They shouldn't even be involved, so removing them works around this bug.
Your fuselage is not self watertight. Both the nose and tail end are open. You will want to close them.
While you can work with thin surfaces all around, VSPAERO works with mixed thick and thin surfaces. For this system, I would use Thick for the Fuselage and Thin for all the lifting surfaces.
Your wing spanwise resolution is not sufficient to get a result you will be happy with.
Your winglet and wing TE are perfectly aligned. This really shouldn't be a problem, but it is causing OpenVSP to detect the winglet wake tracing onto the wing wake -- and then the wing wake is incomplete. I will have to look into this, but the easy workaround is probably to just make the winglet chord slightly shorter so the TE's don't perfectly align.
Rob