Download image fallback after failing to extract with all filters

[danielprosser1]

When all filters are unable to extract an image (including the Webpage filter when images on the page have no size declared), ImageResolver would ideally download the image in order to calculate the size. An example URL with this issue is...

http://www.theguardian.com/media/2016/jun/19/angry-julian-assange-starts-fifth-year-living-in-ecuadors-london-embassy#img-1

Where ImageResolver returns unauthorised link...

https://i.guim.co.uk/img/media/ae7e87eb004579d05ff1b6036ad9390206fbf911/0_51_3500_2100/3500.jpg?w=1200&q=55&auto=format&usm=12&fit=max&s=a18c27165a874395744bf8501d8c2c41

(from OpenGraph filter that the guardian declared for facebook/twitter).

And the correct link (which I had to get manually by dragging the image onto my browser's address bar) is....

https://i.guim.co.uk/img/media/ae7e87eb004579d05ff1b6036ad9390206fbf911/0_51_3500_2100/master/3500.jpg?w=1920&q=55&auto=format&usm=12&fit=max&s=a3d2cb1a9a9685a5bb3eddeb990a9bc7