[mangadex] use 'api.mangadex.org' as default API server

The caching issues seem to be gone.
(#1290, #1310)

gallery_dl/extractor/mangadex.py

@@ -17,6 +17,7 @@ class MangadexExtractor(Extractor):
     """Base class for mangadex extractors"""
     category = "mangadex"
     root = "https://mangadex.org"
+    api_root = "https://api.mangadex.org"
 
     # mangadex-to-iso639-1 codes
     iso639_map = {
@@ -28,7 +29,10 @@ class MangadexExtractor(Extractor):
 
     def __init__(self, match):
         Extractor.__init__(self, match)
-        self.api_root = self.config("api-server") or "https://mangadex.org/api"
+
+        server = self.config("api-server")
+        if server is not None:
+            self.api_root = server.rstrip("/")
 
     def chapter_data(self, chapter_id):
         """Request API results for 'chapter_id'"""