Bug in extracted images sources returning a base64

## Solution
Added by @martintoreilly
1. (minimal fix): update our local copy of Readability.js to the latest file from the Mozilla package
2. (better): stop duplicating Readability.js as a local copy and instead reference the Mozilla package directly (but double check why we opted for a local copy first)

## Original issue
Raised by @cayolblake
Hello,

Any idea why original tag on site looks like the following.

`
`

And the extracted one looks like the following with the "src" attribute messed up?

``

Basically, `src="//cdn.cnn.com/cnnnext/dam/assets/190714161615-ursula-von-der-leyen-children-exlarge-169.jpg" ` ends up being `src="image/gif;base64,R0lGODlhEAAJAJEAAAAAAP///////wAAACH5BAEAAAIALAAAAAAQAAkAAAIKlI+py+0Po5yUFQA7" `

What to do to avoid such behavior?