- It can throw exceptions
- It doesn't return false on things which are not Number, just on things
  that are not NaN

Fixes: https://github.com/jitsi/lib-jitsi-meet/issues/2743

It doesn't get translated in the TS build, for one.

Script I used:

```python
import os

for (dirpath, dirnames, filenames) in os.walk('.'):
    if '.git' in dirpath:
        continue
    if 'node_modules' in dirpath:
        continue
    if 'dist' in dirpath:
        continue
    if 'types' in dirpath:
        continue
    for filename in filenames:
        path = os.path.join(dirpath, filename)
        if not path.endswith('.js') and not path.endswith('.ts'):
            continue
        #print(path)
        with open(path, 'r+') as f:
            #print(f)
            data = f.read()
            if '__filename' in data:
                p, ext = os.path.splitext(path)
                txt = f"'{p[2:]}'"
                print(txt)
                data = data.replace('__filename', txt)  # Assign the result back to data
                f.seek(0)
                f.write(data)
                f.truncate()
```

Waits to check if lobby is already enabled, and if enabled ignore the request.

* fix(events): unify events to a single EventManager type, add support for single typescript declaration

* fix(lint): fix lint

* fix(events): fix incorrect instatiation

* fix(events): clean up redundant methods

* fix(events): keep EventEmitter name, alias NodeEventEmitter

* fix(events): fix loose reference

* fix(EventEmitter): remove on/off alias as redundant

* fix(RTCUtils): bring event handlers under class to use same event emitter

* fix(RTCUtils): fix lint

All modern browsers implement window.addEventListener for the 'error'
and 'unhandledrejection' listeners. Thus there is zero benefit from
using the legacy `onerror` handler here and rolling multiple handlers
ourselves.

There is an old config option `enableWindowOnErrorHandler` which was
never whitelisted in Jitsi Meet, so I'm 100% sure nobody is using it.

Even if someone was, migrarting to the above is simply better in every
possible way.

They should never be needed if the error is handled, and if it isn't, it
will bubble up and end up in the handler anyway due to window.onerror.

It creates its own emitter now extending Listenable.

It's the same as async, but with ESM which is more bundle friendly.

Pull initial offer/answer tasks out of the modification queue and execute them right away. Only track and codec related operations that necessitate a renegotiation cycle need to be pushed to the modification queue. The queue execution is paused until the session is established. This avoids track operations being executed before the session is established. This fixes an issue seen in p2p connections where sources of the initiator are not signaled to the remote since the tracks are added while the initiator is waiting for a session-accept from the peer.

This the second stage in our E2EE journey.

Instead of using a single pre-shared passphrase for deriving the key used for
E2EE, we now establish a secure E2EE communication channel amongst peers.

This channel is implemented using libolm, using XMPP groupchat or JVB channels
as the transport.

Once the secure E2EE channel has been established each participant will generate
a random 32 byte key and exchange it over this channel.

Keys are rotated (well, just re-created at the moment) when a participant joins
or leaves.

Simplifies the logic for delayed ICE failed event by relying
on XMPP ping instead of the networkInfo module to check
if the internet is online.

Add a performance stat around long tasks. Chrome supports PerformanceObserver API that lets us
register for long tasks event. Any task that takes longer than 50ms is considered a long task.

A new method which not only has Promise interface,
but also allows to queue IQs regardless of the current
connection's status.

Ping and send ICE failed notification requests are changed
to use this method which makes it more reliable on XMPP
Websocket connections on mobile(which in contrary to BOSH
dies much faster there).

It makes it easier to remove listeners.

* feat: implement vad noise detection

* start detection services on initial track add

* additional refactor / calibrate noise detection

* stop detection services on track remove

* address code review

* address code review p2

Also do not execute any peerconnection callbacks like
'oniceconnectionstatechange', 'onicecandidate' etc. after close.

* revert: Reverts be665cbff7.

* ref: Renames "peerjid".

* ref: Refactors the initialization of a peer connection.

* feat: Re-implements the A/B test for the "suspend video" feature.

* squash: Deep copy.

* ref: Renames forceSuspendVideo to abtestSuspendVideo.