Move FileDownloader to its own module and create a new class for each download process

A suitable downloader can be found using the 'get_suitable_downloader' function.

Each subclass implements 'real_download', for downloading an info dict you call the 'download' method, which first checks if the video has already been downloaded

setup.py

@@ -71,7 +71,7 @@ setup(
     author_email='ytdl@yt-dl.org',
     maintainer='Philipp Hagemeister',
     maintainer_email='phihag@phihag.de',
-    packages=['youtube_dl', 'youtube_dl.extractor'],
+    packages=['youtube_dl', 'youtube_dl.extractor', 'youtube_dl.downloader'],
 
     # Provokes warning on most systems (why?!)
     # test_suite = 'nose.collector',