The Win7 mingw reftests seem to have frequent hangs, which seem somehow exacerbated by the patch that we want to land for bug 1611660. Per https://bugzilla.mozilla.org/show_bug.cgi?id=1611660#c44 I'm going to disable those tests as they are for an unsupported platform and are blocking work needed for supported platforms.

I did do a fair amount of investigation into the hangs via try pushes with printf logging, and discovered while there are multiple manifestations, they are all caused by the first content process hanging in a Windows file I/O API call. See dep bugs 1642315, 1642331, and 1642363 for the three instances that I chased all the way down. There are more that occur after the content process completes its startup and starts running its event loop, but those are much more involved to track down via printf-logging and I didn't bother.

I'm filing this bug to track down these hang problems, and once those are fixed, re-enable the reftest suite.