Coding vs. Clniical
As a coder, it is not our position to question the clinical decision making of the provider. It is our job use our years of coding education and training to to code what is documented and let the provider, with their years of clinical education and training, decide which conditions are related and which are not.
The code choice E11.628 is an NEC code that aligns with what you state is documented. To be clear though, since these are not normal diabetic manifestations and are not listed under the "with" choices in the alphabetic index, the provider would need to document the cause-and-effect relationship between diabetes and the onchomycosis/callus with appropriate linking terms (with, associated with, due to, etc.) in order for them to be coded to E11.628. Without the cause-and-effect relationship documented, the should be coded as 3 unrelated conditions.