Filter/Ignore Link Targets that point to binary content

In testing, there have been occurances of link target URLs that point to binary content on the remote server (like PDFs, Powerpoint, etc..).

Currently, the HTTPClient used by the application does not discriminate between these links and links to text/html content, and will download and (unsuccessfully) parse the content. By configuring the HTTPClient to identify binary server responses, we can skip downloading these content types, and save report processing time taken to download them. This effort should be included in the performance improvement activity.