Feature wish: fail count in the top summary

Due errors or restrictions in a fuzzer not all benchmark targets are completed by fuzzer. this is only visible when opening up the "Median code-coverages on each benchmark" tab, and if there are a lot of fuzzers and that one is far to the right it might even be outside of the screen.

There is space in the table to the right of the "average ..." value, this could display e.g. 11/21.

I had it a few times that a variant was doing very well or very badly and the truth was it was failing for many experiments