Check whether we get the correct format of the ontology and it can be parsed correctly

[kuefmz]

ideas here are to improve the failover mode

• inspect first lines only to get whether indicator for format (see. e.g. rdf detector)
• parse entire file first (needs performance optimizations e.g. bytesize limit)
• think about caching parse and content type information based on etag or (modified, byte-size, content-type) combo

NOTE: in general this could be also outsourced to archivo itelf for performance reasons since it check this in the crawls anyhow. the report of archivo could be improved such that information would only be fetched from archivo, however this would not be "live" but probably good tradeoff between reliability and speed for full file validity => new mode originalFailoverArchivoMonitor
if these reports are cached in the timemachine and kept in sync then there would be almost no performance effects