syz-cluster: consider global/static variable values

[a-nogikh]

When determining whether a patch series is worth fuzzing, consider not only the hashes of .text symbols, but also the hashes of the global (static and non-static) variables.

As before, calculate the hashes during build and process them at the beginning of the fuzz step.

[dvyukov]

How much is this different from the old existing hashes of all sections? .data was the main difference. There are things like relocations, but these either don't change if nothing in .text/.data has changed, or if changed can have the same effect as changes in .data.

[a-nogikh]

Yes, that needs to checked now.

We anyway need to determine the exact differing .text symbols because these are used to direct the fuzzer, but elfBinarySignature can now probably be reused instead of querying the individual changed symbols.

For .text it's also still important to calculate the hashes on vmlinux.o rather than vmlinux because, if even one symbol differs, there's a chance that, once the call sites are patched with the final PC values, many more symbols will differ as well.