Something is not right, but you knew that already. If the Yamaha inverter generators have the same degree of output isolation as the Honda's, it should be impossible to cause a GFCI trip with a single ground fault, even if you try. If the GFCI itself isn't bad, I would think that there has to be a problem downstream of the GFCI. GFCIs trip from hot-neutral imbalances or neutral-ground leakages. If everything upstream of the GFCI is working correctly it won't be possible to create a hot-neutral imbalance with a single downstream fault. A neutral-ground leakage I would expect would still cause a trip, so I'd suspect that first because that's the only single fault that would cause the trip. If there were leakage from one of the power outputs to ground in the generator or adapter, then a hot-neutral imbalance downstream of the GFCI would cause a trip. Not being a hard failure, this one might be pretty "interesting" to zero in on.