To be a bit more specific, it's computed along the
reciprocal C direction, yes.
Since the K point lies at (1/3,1/3,0) in the Brillouin zone, you should cleave the crystal with Miller indices 110 to get the C axis to point in the correct direction. This is easy to do in the Builder, you just insert graphene from the Database (possibly graphite, for bilayer) and then use "Builder>Surface (Cleaver)".
Now the complex band structure will be computed in a supercell where G-K lies on kC. The supercell is a bit bigger so actually K will be a point on "G-Z", not an end-point, but that doesn't matter.
A full scripts is attached for the calculation - I then took the scripts from the
complex band structure tutorial to make nice 2D and 3D plots, also attached. The whole calculation takes about 30 seconds using a Slater-Koster sp3d5s* basis set and 5001 energy points in the complex band structure.