If anything works, with the shore power turned off, then the bank is grounded.
But maybe not grounded properly...
If loads are grounded to the chassis, then the end of the shunt away from the battery should also be grounded to the chassis. The object is to have all electrons to/from the battery flowing through the shunt.
As it is, that should be happening, but...
Um...it's a bit of a myth that electricity follows the path of least resistance. It will actually follow all available paths. Of course, more will flow where the resistance is least.
I see a possibility of electrons flowing through the chassis getting back to the battery via a non-optimal path. Like through something else...like the Iota or the meter, and then to the ground at the shunt.
Adding a nice fat ground wire from the chassis to the shunt should provide an optimum path back to the battery.
The lack of which could theoretically be causing whacky meter readings...