run_selenium.py error reporting improvements if the server is not running, but with batch mode

[efortuna]

John,
run_selenium.py prints a helpful little error message if you don't have the selenium jar running and you should. It doesn't seem to show up in batch mode though. Can you forward that message on to the user even in the batch case?

[jmesserly]

I think this is because test.dart doesn't print stdout/stderr unless the test fails. Perhaps we should always log stderr?

[efortuna]

But the test should fail if the server isn't running...?

[jmesserly]

I think this was broken by the _ensureSeleniumRunning change. If I add this logging:

  void _ensureSeleniumServerRunning() {
    print('!!! _ensureSeleniumServerRunning avail:$_isSeleniumAvailable starting:$_startingServer');

and run:
  ./tools/test.py -v -n -m release -c webdriver -b safari corelib

I see a bunch of lines like this:
!!! _ensureSeleniumServerRunning avail:false starting:true

but test.dart exits without any other useful messages or running tests. I might be able to take a look later today, but if you want to give it a shot feel free :)

---

cc @efortuna.

[jmesserly]

Added AssumedStale label.