It appears Skywater deleted the W/L parameters on purpose. If you follow all of the include files you end up at the subckt call identified in the folder with the test data and plots. Of course the subckt call expects an l and w to be passed, which was deleted (excluded?) from the test data .spice file card.

I think you are referring to the fixed-layout devices. Since those layouts are fixed, the subcircuit model calls the device model with a given W and L, so the subcircuit no longer takes W and L as a parameter. The test data and plots in the PDK repository are something I did, not SkyWater. The purpose was to check that we didn't break things with the conversion scripts. I intend to use the same/similar checks for the PDK continuous integration system that we will hopefully be working on this week.