As part of bug 687467, our two Mac Tester machines in chromium.webrtc and chromium.webrtc.fyi were upgraded to 10.12. This caused runhooks starting to fail, most likely because it made them start using a clean checkout that didn't previously have anything downloaded.

Example failures:
https://build.chromium.org/p/chromium.webrtc.fyi/builders/Mac%20Tester/builds/39880
https://build.chromium.org/p/chromium.webrtc/builders/Mac%20Tester/builds/68237

It's either Mercurial and/or the AppRTC build that fails:
________ running '/usr/bin/python webrtc.DEPS/build_mercurial_local.py' in '/b/c/b/Mac_Tester'
python setup.py  \
	  build_py -c -d . \
	  build_ext  -i \
	  build_hgexe  -i \
	  build_mo
running build_py
running build_ext
running build_hgexe
running build_mo
warning: hgbuildmo: could not find msgfmt executable, no translations will be built

env HGRCPATH= python hg version
Mercurial Distributed SCM (version 3.2.2)
(see http://mercurial.selenic.com for more information)

Copyright (C) 2005-2014 Matt Mackall and others
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

________ running '/usr/bin/python webrtc.DEPS/build_apprtc_collider.py' in '/b/c/b/Mac_Tester'
failed MSpanList_Insert 0x74c178 0x468e4396214 0x0
fatal error: MSpanList_Insert

Since we've recently done work in moving this to the WebRTC repo, we should replace these runhooks steps with a separate buildbot step calling the download and setup scripts in https://chromium.googlesource.com/external/webrtc/+/master/webrtc/tools/testing
(living in src/third_party/webrtc/tools/testing of a Chromium checkout).

This bug tracks that work. Mirko: let's work together on this on Monday.
CCing upcoming WebRTC-in-Chrome sheriffs.