Pcfinhom and Kcross.inhom unexpected behavior, heterogeneity or not?

I am currently analyzing the spatial patterns of a tree species using the " spatstat " package - point pattern analysis (pcf, pcfinhom, pcfcross.inhom, Kcross.inhom).
I struggle with heterogeneity and bandwidth (selection) concepts and functions output (strange graphs).
I posted the data and the questions here:
I would be very grateful to receive some advice.
Thank you in advance!
Best regards,