test: add known issue test for child_process.spawnSync segfault

[imyller]

Checklist

• make -j4 test (UNIX), or vcbuild test nosign (Windows) passes
• tests and/or benchmarks are included
• commit message follows commit guidelines

Affected core subsystem(s)

test

Description of change

child_process.spawnSync segfaults when given an array/object for the file argument with a throwing toString definition.

This PR adds a known_issues test with expected crash result.

Ref: #8539

[bnoordhuis]

I don't see the point. You can crash node.js through almost any native function that accepts non-primitive values, see #7902; it's certainly not exclusive to spawnSync().

[imyller]

@bnoordhuis Agreed when that is the case.

From end-user standpoint issues #8537, #8538 and #8539 fall in to same category and should be explained and resolved gracefully.

I opened this PR to match #8539 to allow easier testing if per-case solution for core lib segfaults is later added.

[bnoordhuis]

From end-user standpoint issues #8537, #8538 and #8539 fall in to same category and should be explained and resolved gracefully.

#7902 (comment) is the explanation including how to fix it. But as I said there, changing everything to v8::MaybeLocal<T> is not going to happen overnight; it's on my for-when-I-have-a-spare-week list.

[imyller]

@bnoordhuis Thank you for the explanation.

I'll close this PR as it does not add any additional value to the already well known issue.