Make SDP such that h264 is negotiated with Firefox.

[stefhak]

However, I don't think it should be merged since it seems quite unstable, it seems owr often stops decoding after a short while (leaving a frozen video). Output in console (where openwebrtc-daemon is run):

(<unknown>:50476): libnice-WARNING **: Could not find component 2 in stream 1

** (<unknown>:50476): WARNING **: (owr_transport_agent.c:2414):on_new_remote_candidate: runtime check failed: (cands_added > 0)

(<unknown>:50476): libnice-WARNING **: Could not find component 2 in stream 1

** (<unknown>:50476): WARNING **: (owr_transport_agent.c:2414):on_new_remote_candidate: runtime check failed: (cands_added > 0)

(<unknown>:50476): libnice-WARNING **: Could not find component 2 in stream 1

** (<unknown>:50476): WARNING **: (owr_transport_agent.c:2414):on_new_remote_candidate: runtime check failed: (cands_added > 0)

(<unknown>:50476): GStreamer-CRITICAL **: gst_object_unref: assertion '((GObject *) object)->ref_count > 0' failed

(<unknown>:50476): GStreamer-CRITICAL **: gst_object_unref: assertion '((GObject *) object)->ref_count > 0' failed

(<unknown>:50476): GStreamer-CRITICAL **: gst_object_unref: assertion '((GObject *) object)->ref_count > 0' failed

[stefhak]

Closing since this was apparently already merged. However, in my experience this gives issues (often the FF open264 plugin crashes) so I'm not sure it was a good move to merge.

[superdump]

We should really expose the actual profile/level based on caps rather than hardcoding something.

[stefhak]

Agree, and given how often FF crashes when h264 is negotiated with this fix I think we should undo it (we're better off using VP8 for the time being).