Investigate flaky test test-child-process-spawnsync

[joaocgreis]

Examples of failures:

• centos7-64

[Trott]

If I'm understanding the test correctly, the part of the test that is flaky is actually invalid. It checks to see if the function passed to setTimeout() executes in the same second as call to spawnSync() finishes. This will not always necessarily be the case, and the test will only get flakier if more things are added to it. (There are two other test cases below the flaky one.)

There's already an assertion to confirm that the spawnSync() ends before setTimeout() runs, so I'll submit a PR to remove the check altogether rather than do something like convert the timestamps to nanoseconds since start and make sure the one set by setTimeout() is greater than the one set by spawnSync().

[Trott]

Fixed in 7e63eb7 / #2535