Test app tests

I'm currently manually validating the test app on Mac OS and Linux, but it's built to be tested in an automated fashion. On each OS, we should launch the app, send a signal, and validate the output contains "HOOK". This should be done for the native version of the app and the JVM version on each OS.

On POSIX this is relatively easy since we have access to PIDs and `kill` API. On Windows we can only be signaled with Ctrl+C so I think you just attach a console and then there's an API to send that event.

Start by locating the test app and the existing native and JVM test entry points, then read how each variant is launched on the supported operating systems. Implement automated runs that signal each app and verify output containing "HOOK"; done means both variants are covered on each OS, including the platform-specific signaling behavior.

Written by the indexing model from the issue text.