Moves the logic from all different places into single state
listener to combine all inputs into a single output.

Also, use the configured resolution to set it as the max received frame size.

Up until now we relied on implicit loading of middlewares and reducers, through
having imports in each feature's index.js.

This leads to many complex import cycles which result in (sometimes) hard to fix
bugs in addition to (often) breaking mobile because a web-only feature gets
imported on mobile too, thanks to the implicit loading.

This PR changes that to make the process explicit. Both middlewares and reducers
are imported in a single place, the app entrypoint. They have been divided into
3 categories: any, web and native, which represent each of the platforms
  respectively.

Ideally no feature should have an index.js exporting actions, action types and
components, but that's a larger ordeal, so this is just the first step in
getting there. In order to both set example and avoid large cycles the app
feature has been refactored to not have an idex.js itself.

When the reccorder joins, they have a local participant, which is not rendered,
so don't count it towards the partcipant count used for computing the tile
sizes.

Since the verical filmstrip doesn't set its width explicitly anymore,
calculating the available area for the large video based on the
filmstrip width retrieved from the HTML element was wrong
in the cases when the rendering and cleanup of the filmstrip hasn't
finish yet. For example when switching from tile view to stage view.

* ref(Filmstrip): Optimize resizes.

* fix(thumbnails): resize.

* fix(thumbnails): Issue with height 0, width 0.

* doc(Filmstrip): Improve JSDoc.

Any overrides set on interfaceConfig are not
applied on module load. As such, call to get
the value of the auto pin setting, providing
time for the bootstrapping to set overrides.
Otherwise iframe api users cannot override
the setting.

React Native doesn't define __filename nor __dirname so do it artisanally. In
addition, this helps with centralizing the configuration passed to loggers.

The store's remote tracks get cleared
and re-added when switching in and out
of p2p, which can cause pinning thrashing.
Avoid that with a debounce.

Use subscribers to detect state change and emit those
out to other participants. Use middleware to register
the command listener.

🔥🔥🔥

Using anything non-serializable for action types is discouraged:
https://redux.js.org/faq/actions#actions

In fact, this is the Flow definition for dispatching actions:

declare export type DispatchAPI<A> = (action: A) => A;
declare export type Dispatch<A: { type: $Subtype<string> }> = DispatchAPI<A>;

Note how the `type` field is defined as a subtype of string, which Symbol isn’t.

- Reset some state on the singletons conference
  and VideoLayout.
- Add a way for LocalVideo to clean itself up
  by sharing logic with the other SmallVideos.
- Add clearing of chat messages so they don't
  linger.
- Remove some UI event listeners.

* feat(tile-view): initial implementation for mobile

- Create a tile view component for displaying thumbnails in a
  two-dimensional grid.
- Update the existing TileViewButton so it shows a label in the
  overflow menu.
- Modify conference so it can display TileView while hiding
  Filmstrip.
- Modify Thumbnail so its width/height can be set and to prevent
  pinning while in tile view mode.

* use style array for thumbnail styles

* change ternary to math.min for expressiveness

* use dimensiondetector

* pass explicit disableTint prop

* use makeAspectRatioAware instead of aspectRatio prop

* update docs

* fix docs again (fix laziest copy/paste job I've ever done)

* large-video: rename onPress prop to onClick

* change forEach to for...of

* use truthy check fallthrough logic instead of explicit if

* put tile view button second to last in menu

* move spacer to a constant

* the magical incantation to make flow shut up

* feat(tile-view): exit tile view on pin

* Try out this other ux impl

* feat(tile-view): persist setting in local storage

* comment

* feat(tile-view): initial implementation for tile view

- Modify the classname on the app root so layout can adjust
  depending on the desired layout mode--vertical filmstrip,
  horizontal filmstrip, and tile view.
- Create a button for toggling tile view.
- Add a StateListenerRegistry to automatically update the
  selected participant and max receiver frame height on tile
  view toggle.
- Rezise thumbnails when switching in and out of tile view.
- Move the local video when switching in and out of tile view.
- Update reactified pieces of thumbnails when switching in and
  out of tile view.
- Cap the max receiver video quality in tile view based on tile
  size.
- Use CSS to hide UI components that should not display in tile
  view.
- Signal follow me changes.

* change local video id for tests

* change approach: leverage more css

* squash: fix some formatting

* squash: prevent pinning, hide pin border in tile view

* squash: change logic for maxReceiverQuality due to sidestepping resizing logic

* squash: fix typo, columns configurable, remove unused constants

* squash: resize with js again

* squash: use yana's math for calculating tile size

- Instead of having 4 listeners for local connection status
  updates and 1 for remote, remove two of the redundant listeners.
- Instead of calling into 4 separate VideoLayout methods to update a
  participant's connection status, expose one handler.

* ref(video-layout): move middleware for TRACK_ADDED

* ref(video-layout): call mucJoined when redux knowns of conference join

Instead of keeping dominant speaker locally, get it from redux and be
updated when the dominant speaker changes. This is in an attempt to mimic
the video layout being reactified and connected to redux.

The goal will be to make video layout stateless and instead
get all state from redux.