Add boundscheck in speccache_eq to avoid OOB access due to data race

[kpamnany]

Like #54671, but for speccache_eq.

Saw another segfault with this in the stack trace, hence this fix. I also looked for other uses of jl_smallintset_lookup and there's one in idset.c. That doesn't appear to be racy but I'm not familiar with the code, so maybe you can take a look at it in case we need to push a fix for that one too @gbaraldi or @vtjnash?

[vtjnash]

The idset_eq probably has the same issue, since it could be used in a race-y way by users, and it is supposed to be safe (if not necessarily defined behavior) in that case

[kpamnany]

idset.c isn't in 1.10.2 which is our current version. Also, I didn't see a similar pattern in idset_eq (of using an svec racily) so I'm not quite sure what the race would be there.