Fix/re-enable scraper for link topic favicons
Link topics are supposed to have their favicons scraped automatically, but there were some issues with the scraping process so I disabled it and haven't yet gotten back around to fixing it up. That needs to be done before too much longer though. Initially it should probably just handle the favicons, but in the future I can look at adding some of the other planned scraping behavior (like collecting info about when the linked page was published, potentially who the author is, and so on).