The server serves malformed header "Content Type: text/xml" for the XML
request (it should be Content-Type but not Content Type). Python 3.x,
which uses email.feedparser rejects such headers. As a result,
Content-Encoding header is not parsed, so the returned content is kept
not decompressed, and thus XML parsing error.
Also, m3u8_url extraction should be fatal as all formats depends
directly or indirectly on it.
This change fixes test_Generic_26 and TestFunnyOrDieSubtitles