ESLint 4.8.0 discovers a lot of error related to formatting. While I
tried to fix as many of them as possible, a portion of them actually go
against our coding style. In such a case, I've disabled the indent rule
which effectively leaves it as it was before ESLint 4.8.0.

Additionally, remove jshint because it's becoming a nuisance with its
lack of understanding of ES2015+.

The counterpart of the external API in the Jitsi Meet Web app uses the
search URL param jwt to heuristically detect that the Web app is very
likely embedded (as an iframe) and, consequently, needs to forcefully
enable itself. It was looking at whether there was a JSON Web Token
(JWT) but that logic got broken when the JWT support was rewritten
because the check started happening before the search URL param jwt was
parsed.

Fixes not able to close it with esc button.

Removing translateString forces using data-i18n attributes, to make sure we do not forget to set them. Missing data-i18n attributes is a problem with late loading where we can end up without translation, without text. Missing data-i18n attributes is also problem that strings will not be translated when changing language.
Fixes a bug in invite dialog, where remove password button was shown for non moderators.

Adds more analytics events for shortcuts, recording and shared video. Changes the way we calculate ttfm.

Use KeyboardEvent.key if available,
match both lower and upper case letters to keep previous behaviour

KeyboardEvent is a mess.

KeyboardEvent.which gives you, in theory, a decimal representation of the key pressed.
"r" or "R" gives you 82, which is "R", you can look at KeyboardEvent.shiftKey,
but you don't have access to capslock...

Maybe you want to use numbers, but of course NumPad will not give you the same than
"normal" numbers ...

Now if you use something else than letter, for exemple "?",
on a QWERTY keyboard "/" and "?" gives you 191,
on a AZERTY keyboard "," and "?" gives you 188, so we have to stick to letters.

This was for keydown and keyup events, keypressed event return the real char
(lower "a", "/", "?", ...) but it fails in some cases

The only non broken property is KeyboardEvent.key,
but it's only supported since Chrome 51, Opera 38, and not supported by Safari
https://developer.mozilla.org/en-US/docs/Web/API/KeyboardEvent/key

A good reference & test:
http://unixpapa.com/js/testkey.html
http://unixpapa.com/js/key.html

Signed-off-by: Etienne CHAMPETIER <champetier.etienne@gmail.com>

This reverts commit 93648f361b.