Prevent faulting by removing logs that use unmapped logging structures

[JDuchniewicz]

Since refactoring the whole test suite is too big of a task for the time being - I just removed the offending logs informing about page unmapping (and page table) failures that resulted in faults described in #29.

[midnightveil]

Surely these unmaps shouldn't be failing at all?

(Also, did doing a full instead of a "shallow" clone for the test process not work? it should be a single function call difference...?)

[Indanz]

Also, if bench_proc being a separate process complicates things, perhaps just run the code in the main thread instead?

I suspect that the unmap fails because of the vka_cnode_revoke(&untyped_path); call in main(). If bench_proc has the same priority as the main thread then the main thread will do that after receiving the result. If this is what happens, the unmap calls can be just removed, as no further cleanup is necessary.

[Indanz]

When encountering problems, always try to find the root cause and fix that instead of only the symptoms (though symptoms can also be improved often enough).