It looks like something was off in your test:
toolhead: X:130.000000 Y:140.000000 Z:5.001502 E:0.000000
gcode: X:130.000000 Y:140.000000 Z:5.000000 E:0.000000
and:
toolhead: X:130.000000 Y:140.000000 Z:4.999663 E:0.000000
gcode: X:130.000000 Y:140.000000 Z:5.000000 E:0.000000
Note the discrepancy between the requested gcode position and the requested toolhead position. So, you didn’t make it back to the same starting position when you ran the second get_position. Sounds like a bed_mesh, z_thermal_adjust, or something like that was in effect.
Also, I would recommend changing the macro to go to an explicit XYZ position (not just Z). See [BUG] Accumulating microstep shift during probe moves related to endstop oversampling - #3 by koconnor
FWIW, your logs look like the patch was successful for you, as the initial report was a relative change in Z stepper locations, where as now you have an absolute change, which appears to be the result of some gcode transform in effect
Cheers,
-Kevin