Yeah, much of that is related to people using XLRs for line-level connections between gear on different circuits. I certainly hope that damage to gear is not at issue, but there is some crazy sloppy stuff out there, especially older stuff. But usually people are just worried about interference from ground loop current--which should never happen in balanced gear, but again that's a design issue.
For mics it's much simpler; the mic's chassis will be bonded to its pin 1, period. The main issue there is how good that bond is. This is why the Neutrik EMCs can help, the bond in their female connector can be better at RF than the mic's internal bond. So you can do whatever you want in the female connector, you can't make it worse I don't think, and you might make it better.
On the male end, again, pin 1 *should* be bonded to the chassis at the female pin 1 input to the device. If not, you can do so in the male connector if you like. It is becoming more common for modern gear to be properly wired, so I suspect it doesn't make much difference. It's easy enough to test on gear with a metal chassis: connect a shell to the input and see if there is continuity between the shell and the chassis. But on a plastic-body recorder, who knows really? Even a plastic-body XLR connector can have a grounding tab for the shell to mate with, but you can't measure that directly if the case itself is plastic--it can be internally shielded instead.
My preference is to bond all of my cable shells, but I am working entirely with gear I built or modified as necessary.