You've made many of those meters jump around and vaporize copper without resorting to the current input jacks.
Fungus, I think that the blown fuse test that they are required to run considers that the rest of that area isn't going to breakdown (arc over) under normal conditions. This assumes the fuse contains the arc.
Indeed, I have made a few of them jump just as you suggest, using the voltage inputs. The small generators put about about 20J. That in itself isn't enough to make the meters jump but couple it with 4-600 Joules, we can even split a case open.
Most meters won't break down (or arc over) internally with the voltage inputs at the 6KV I test at. So I could have 100KJ available and it would not matter. All the energy just dissipates internal to the generators output stage in heat.
Of course, when the meters have these cheap fuses in them, I look at it as an opportunity to play with the current inputs as well.
A bit off topic.
The hobbyist working in CAT I most likely doesn't care about any of this HRC stuff. I imagine most are working on their bench with very low energy circuits and chances of an arc flash are zero. Then we have my favorite, avg ave canadian guy that made a rant video about meter safety all the while with showing they had blow a fuse and not replaced it with the proper safety fuse.
Again, with some basic education, you could get away with a lot to the point of bypassing the fuses all together (which is what I typically do on the bench when I am using shunts). At home, I fit this hobbyist group. I don't care about HRC fuses and really, don't care about using the meters to directly read current. But, I also don't want a meter that when I do something stupid, that the meter is damaged. For me, it's not ever been a concern for safety but robustness. IEC 61010, who cares. 61326 is where it's at. Again
When I work in CAT III, I use a meter with clamps. There is no fuses. And, the meter is certified.