Following a suggestion by Asger, we track use nodes through calls to `promisify`. When we see a call to a promisified function, we introduce a new synthetic API-graph node representing the callback argument synthesised by the promisification, and track the result of the call to an `await` (or other promise resolution), which is then considered to be a use of the first parameter of the synthetic callback (the zeroth parameter being an error code, which we do not model yet).
With the old test runner we cannot have `VerifyAssertions.qlref`s for each individual test that reference a shared `VerifyAssertions.ql` in the parent directory, since it doesn't like nested tests.
Instead, we have to turn `VerifyAssertions.ql` into `VerifyAssertions.qll`, and each `VerifyAsssertions.qlref` into a `VerifyAssertions.ql` that imports it.
But then that doesn't work with our old directory structure, since the import path would have to contain the invalid identifier `library-tests`. As a workaround, I have moved the API graph tests into a directory without dashes in its path.
