Keep demultiplexing STUN/RTP/RTCP packets after DTLS ones.
We might end up processing the session a little longer, because we will
process the STUN/RTP/RTCP packets after the DTLS handshake.
Try to have a faster classification, on first packet; use standard extra
dissection data path for sub-classification, metadata extraction and
monitoring.
STUN caches:
* use the proper confidence value
* lookup into the caches only once per flow, after having found a proper
STUN classification
Add identification of Telegram VoIP calls.