Suicide SHOULD be included in gun deaths. So you're saying that in states like AZ, WY, and AK, everyone is just committing suicide there and not murdering anyone? It's not a deliberate false chart, you just don't like it because it shows the truth.
Probably best to keep suicide and homicide figures distinct, though. Both are important, both are affected by availability of guns, but nevertheless they are different
moral issues. People tend to be accepting of much stronger restrictions to avoid the latter than the former. Hence even countries that ban guns don't ban over-the-counter medications that can be used for suicide, they just restrict them slightly.
(Same thing with national self-harm. The rest of the world has a lot more moral and practical incentive to respond forcefully to countries that attack others than to countries that have collective mental-health issues causing them to decide to injure themselves with weird political choices. Speaking purely hypothetically.)
Does an armed teacher have a better chance to protect their kids in a shooting event than an unarmed teacher?
Not so much of a difference as to justify the downsides, I'd say. For the armed teacher to make a difference there would have to be an armed teacher there that one time a shooter goes on a spree in that particular school (not all teachers would be prepared to be armed, and in some cases the one armed teacher might be dropped from a distance before they even know the shooter is there, they might be the first one shot) and that teacher would have to actually respond effectively (the majority probably would not, looking at real-world examples).
Set against that would be all the times when something went wrong and the teacher themselves was the one who went nuts, or when the gun goes off due to the teacher's incompetence at gun-handling, or when somehow a student gets hold of the gun. Also set against it has to be the very-hard-to-quantify effects of taking such a further step to normalise guns and further expand the lobby against restricting them.
Weirdly, this reminds me of arguments about bicycle helmet laws. You have to compare practical effects in very narrowly-defined circumstances ('if someone hits you over the head with a steel bar would you rather have a helmet or not?'), versus much harder-to-quantify wider social effects ('what message does such a law send to bad drivers about the burden of responsibility for victims vs perpetrators, and what does it do to cycling rates and hence to rates of heart-disease and pollution-related deaths?'). There's often such an asymmetry between easy-to-grasp effects in very specific situations vs much more distributed large-scale consequences.