new iOS/web extractors, image proxying done by views.py

- ythdd_globals.py - added helper function to get user-configured header
- ythdd.py - now checks for config.toml in work directory
- requirements.txt - add brotli, so that requests can decompress
innertube request

ythdd_api_v1.py

@@ -113,7 +113,7 @@ def hot(data):
 			started = time.time()
 			try:
 				# try to actually get the data
-				extracted_related = ythdd_extractor.related('https://www.youtube.com/watch?v=' + videoId)
+				extracted_related = ythdd_extractor.WEBrelated('https://www.youtube.com/watch?v=' + videoId)
 				extracted_related['took'] = time.time() - started
 				return 200, "OK", extracted_related
 			except KeyError: