Move logic to open device selection outside of SettingsMenu so
it can be called independently by either SettingsMenu or by
the settings button itself if no other settings but devices will
be displayed.

If toggle audio or video is executed too early and the local
tracks don't exist we fail to execute the operation. Now we store
the mute state and we are executing it after the tracks are
created

If toggle SS is executed too early and lib-jitsi-meet is not yet
initialized toggle SS will fail. Now we are storing the whether
SS is on or off and when lib-jitsi-meet is ready we are starting
SS if needed.

Cleanup existing logic for displaying and updating device
selection settings in the settings menu. In its place
is a button to open the device selection modal.

The Device Selection modal consists of:
- DeviceSelection, an overly smart component responsible for
  triggering stream creation and cleanup.
- DeviceSelector for selector elements.
- VideoInputPreview for displaying a video preview.
- AudioInputPreview for displaying a volume meter.
- AudioOutputPreview for a test sound output link.

Store changes include is primarily storing the list of
available devices in redux. Other app state has been left
alone for future refactoring.

* Uses new peer connection statuses to check and show different user msgs.

Checks for interrupted state of peer connection and shows appropriate messages. In case of inactive or restoring state a message is show to user that video was stopped on purpose. Removes some unused parameters from the event handlers about peer connection status change.

* Removes isParticipantConnectionActive.

#1411

Replaces changeAvatarID, changeAvatarURL and changeEmail with
participantUpdated action.
participantUpdated can be fired for local user without id. This
fixes the problem with updating the local user before the user
join the conference which results in fix for failing to execute
commands for avatarID, avatarURL and email right after the iframe
api creates the iframe with Jitsi Meet.

If executeCommand('displayName') is executed before Jitsi Meet
is fully initialized some listeners were not added and the
display name was not changed.

Uses leavingIDs to more efficiently iterate over remote videos.

* Handle last n in the client

* fix(LargeVideoManager.js): Fixes check for low bandwidth. Needs more work

* fix(LargeVideoManager.js): Fixes the Shared Video test.

* fix(LargeVideoManager): Fix shared video view and remove last n checks.

* fix(LargeVideoManager): Fixes jsdoc comment

* fix(RemoteVideo): Fix connection status check

* fix(LargeVideoManager,RemoteVideo): Syntax errors

Also clears debian package from including *.js files from source roote folder, files which were not used.

We have the functions reload and redirect which modify window.location.
Use them and do not directly modify window.location so that we have
fewer places of direct window.location modifications and it is easier to
refactor them.

* M1: device change now uses new flow.  fundamentally "works" but may be corner cases/side effects to other flows. haven't touched ffox yet

* M2: change toggle screenshare flows to use the new video replacement chain

* remove the old 'useVideoStream' and replace it with the new one

* use the new (and renamed back from the shim) 'dispose' method

* tweaks to work with the sdp overhaul changes in lib-jitsi-meet

* change the order in which we call dispose (to handle dispose being reverted back to how it is currently on master)

* move useAudioStream over to new flow

* restore useVideoStream doc

* handle rename JitsiConference::replaceStream -> JitsiConference::replaceTrack

* fix useAudioStream and useVideoStream to return a promise again

Using sessionStorage to save whether user is guest or not.

Also removes device change listener, cause when PC wakeups devices will be reconnected and we can try open the local video again.