No doubt this is desirable. However, adding all the CSS features required to support cgit may have been a lot more work than editing cgit's CSS. It's an attempt at avoiding yak shaving; adding recursive sub-projects that balloon a project's scope of work far beyond the original plan.

Dillo is actively developed, and the project of "migrate away from github" is complete, so now other work can be started and completed (like adding the CSS features required to support mainline cgit).