@User did you ever get to the bottom of it? As far as I can tell, this is the rule I'm seeing for the hd cells:

SYMMETRY y  ;
    CLASS CORE  ;
    SIZE  0.460 BY 2.720 ;

Which your cells abide by. The official SkW cells at https://foss-eda-tools.googlesource.com/skywater-pdk/libs/sky130_fd_sc_hd have

SITE unithd ;

And your cells do as well. Not sure what the issue could be, but I am seeing this earlier in your log file:

Area for cell type \AND2X1 is unknown!
   Area for cell type \AND2X2 is unknown!
   Area for cell type \AOI21X1 is unknown!
   Area for cell type \BUFX2 is unknown!
   Area for cell type \INV is unknown!
   Area for cell type \INVX1 is unknown!
   Area for cell type \INVX2 is unknown!
   Area for cell type \INVX4 is unknown!
   Area for cell type \NOR2X1 is unknown!
   Area for cell type \OR2X1 is unknown!
   Area for cell type \OR2X2 is unknown!

I thought that maybe the issue might be related to the .lib file? But you do have areas in there. The areas at https://raw.githubusercontent.com/thesourcerer8/caravel-stdcelllib-stdcells/release/cells/lib/libresilicon.lib seem to be in the wrong order of magnitude though, just from a quick glance. That's probably not it...

The problem was that I had placed the cells wrong with the macro_placement.