Add Webpage as Context

[harrismcc]

It would be very cool to be able to include webpages as context in my queries. For example. I could toss in a link to some documentation or a README from github. It could be used with a syntax like #https://website.com or something.

I think once the site content is actually scraped (probably the hard part) the existing pipelines for code context could be re-used.