Given a module M, to know all the 'downstream' clients of M you need a system that knows about all the relevant clients. This is only possible by integrating with a central repository, e.g. something like Artifactory that tracks for each project what all of its dependencies are and also maintains a reverse index.
Perhaps it also recompiles and re-tests everything dependent on module M and tells you about any errors, giving you continuous integration.
If you just want to know the dependencies of module M, or view and debug the intermediate computations to produce M, that's a much easier problem because it can be isolated. Though, it would still require a lot of tooling support to effectively reflect on the compilation process.