Hi Steve,
Arcs: there should be no problem with Carbide Motion, as far as I know it does not interfere with G2/G3 commands sent to GRBL (that does the heavy lifting), and for example the Fusion360 post-processor for the Shapeoko uses them and works fine with CM.
There’s always the matter of G2/G3 errors thrown by GRBL when…exotic conditions happen that create rounding problems, but that’s a different story.
The return to the defined zero position should be fine too, but looking at the last lines in the example G-code file you posted:
[…]
X125.245Y20.500Z-0.500
G0X125.245Y20.500Z4.000
G0X0.000Y0.000Z4.000
G0X0.000Y0.000Z5.000
I’m just wondering if 4mm is a clearance value you set in devCad ? (and therefore that you set it to 4mm because you knew there would be no clearance issue for the cutter moving back to 0,0 ?). If so, I don’t see how this could hurt.
If think CC generates a return to “center/back of the machine” as a convenience to clear the way for the user to access the piece, nothing more.