Make sure the codec order is munged on all the m-lines for unified plan clients. Implement the logic for setting the preferences through RTCRtpTransceiver#setCodecPreferences.

Use the new Colibri message "EndpointStats" for broadcasting the local stats. The bridge then will be able to filter the endpoint stats and send them only to the interested parties instead of broadcasting it to all the endpoints in the call.

 RTCRtpReceiver.getCapabilities could be undefined on some platforms

Drop end to end signature. The impersonation attack vector is deemed out of
scope since insiders are already part of the meeting.

Capture the audio levels only for the top 5 speakers as RTCRtpReceiver#getSynchronizationSources can be expensive when we have too many audio receivers in the call.
Also, capture the audio levels for track that are unmuted if RTCRtpReceiver#getSynchronizationSources is not supported.
Switch Safari to using getStats since its reporting errorneous values, i.e., 0.000001 as audio level for all remote audio tracks.

Stop using the legacy pc constraints that are no longer wired up to WebRTC.

* fix(TPC): Configure degradation preference in RTCRtpSendParameters.
Properly configure degradation preference on RTCRtpSendParameters instead of RTCRtpEncodingParameters. Fixes https://github.com/jitsi/lib-jitsi-meet/issues/1510.

* feat(JingleSessionPC): Remove ssrcs from remote desc when a user leaves.
Remove the ssrcs (associated with remote sources) from the remote desc along with the removal of the remote tracks when an endpoint leaves the call. The source-remove signaling message from Jicofo will no longer be needed in this case and can be dropped.

Add the ability to send the bridge messages for the receiver video constraints in the new format directly.

Apple added getUserMedia support for WkWebview based browsers like chrome and Firefox on iOS 14.3. These browsers behave as Safari does on iOS. Therefore, extend the Safari checks to these webkit based browsers as well.

All browsers on iOS use WebKit engine so they use the Unified Plan.

RTCUtils.enumerateDevices was being called earlier than the RTCUtils.init which assigned that function

By looking at the replaceTrack usages it's possible for it to be called with both new and old tracks null/undefined.

When a second remote track of the same mediatype is received for an endpoint, remove the existing track before creating the new remote track.

* Added video mute participant

* Trigger mute event

* Optimized mute type checks

* Fixed event name

* Fixed some linter issues

* Fixed more linter issues

* And even more linter issues fixed

* And more linter fixes

* Added media type to analytics event

When E2EE is disabled we need just set the last key to `false`. When decoding we
need to check if the last key is not false, otherwise we might try to decrypt
the unencrypted video frames, thus resulting in breakage.

- After ratchet do not set it as new key for current index. Set it to correct position in crypto ring.

* feat: Drops unused code about handling incoming caps changes.

* fix: Drop using private fields outside JitsiParticipant.

There were cases where the bridge was sending the same constraint multiple times causing redundant calls to getParameters/setParameters on the RTCRtpSender.

Translate the 'LastNChangedEvent', 'SelectedEndpointsChangedEvent' and 'ReceiverVideoConstraint' messages into the new 'ReceiverVideoConstraints' message that invokes the new bandwidth allocation algorithm in the bridge that is described here - https://github.com/jitsi/jitsi-videobridge/blob/master/doc/allocation.md. useNewBandwidthAllocationStrategy=true in config.js will invoke the translation in the client.

Fire PERMISSION_PROMPT_IS_SHOWN when none of the devices have a label
Fire a new SLOW_GET_USER_MEDIA event if the timeout for getUserMedia is exceeded

Update JitsiMeetJS.createLocalTracks to include the options for firing the above events
in the provided options argument. Deprecate the firePermissionPromptIsShownEvent flag in
method's signature

All the send video constraints for the client, i.e., what simulcast streams will be enabled based on constraints received on the bridge channel, will be handled by the SendVideoController class.
The receive video constraints like lastN, selectEndpoints and receive video resolution will be handled by the ReceiveVideoController class.

* fix: Save guards _features to be always empty and nver undefined.

* fix: Ignore caps changes relaying only on features in presence.

"Pinned" endpoints are deprecated, use "selected" instead.

Make sure screenshare w/ presenter gets the full 2500 Kbps bitrate.