My bottom line take from this exercise is that Gradle driven Kotlin code-coverage is in a pretty sorry state and needs about one person-years worth of TLC to be ready for prime-time. It needs to be Kotlin-centric (whereas JaCoCo is both Java and byte-code centric. In other words, big effort and best done by Breslav
et.al. I hope JetBrains takes on the task, at least in leadership, and focuses on a Gradle solution that IntelliJ can leverage. Something that works for both multiplatform and single platform.