I offered several alternatives and I made it clear they belong in a spectrum. Don't make it sound like I claimed it was binary.
If the options are "Keep writing in Scala", "Port to Kotlin" and "Write a Scala compiler", the latter is clearly the worst and most costly of all.
Porting a large codebase can easily be harder than writing a compiler, many companies have codebases much larger than a reasonable compiler.
ah yes rewriting in a new language, the classic not-insane approach