TLS fingerprint with AG should be the same as without it

[ameshkov]

Here's the problem, currently it's easy to distinguish AG from a "clean" browser judging from the TLS fingerprint.

• Chrome: https://tlsfingerprint.io/id/8466c4390d4bc355
• Chrome with AG: https://tlsfingerprint.io/id/958829c358561708

The difference is simple, Chrome sends one extra TLS extension compared to us: https://tlsfingerprint.io/find/extension/4469

Moreover, ideally, our TLS fingerprint should be the same as the browser's FP, not just as Chrome's.

[ngorskikh]

https://source.chromium.org/chromium/chromium/src/+/cc8598642336ec1fbf0eaf9c226f96b0d0acdaaf
https://datatracker.ietf.org/doc/html/draft-vvv-tls-alps

Chrome has implemented a draft rfc allowing the exchange of HTTP/2 SETTINGS frames as part of the TLS handshake.