Parser issue: finds the wrong `title` element

On pages such as [this Smashing Magazine article](https://www.smashingmagazine.com/the-smashing-newsletter/smashing-newsletter-issue-406/), webmentiond parses the last `title` element it sees, even if that title element is not part of the `head` section. The result is that the page's title is assumed to be "Clear Search", even though that's actually the title of an `svg` document embedded in the page.