If the government, for some reason, decided to only communicate via telephone calls, there would still be no duty on any private phone company to provide anyone with the ability to call the government. The government would implicitly require itself to start up and operate a public telephone company, so that any person unwilling or unable to make phone calls via a private company could use it to communicate with the government instead.
In all likelihood, this would mean that VoIP phone booths would appear in every post office, and only calls from those phones to phones registered to government offices would be connected. They wouldn't even necessarily have to use phone numbers or connect to the PSTN; a .gov or .us SIP registration server would do.
For private services, if the government uses them at all, it must use them in a fair and nondiscriminatory manner. The private company has no duty to enforce this on its government users, or to police their behavior.
> Surely that must mean that Twitter banning people is also a violation of the first amendment
Right, and it's nonsensical to apply the first amendment to Twitter, because, broadly speaking, users don't have standing to make First Amendment claims against a private entity like Twitter.
If Twitter is banning people in a problematic way, or in problematic numbers, then there might be an argument that the government can't use Twitter to publish breaking news or announce policy changes. There are, in fact, separate lawsuits around this. But still, that's a far cry from saying "Twitter banning people is a violation of the First Amendment".
No, because that's not what this ruling does. This ruling says that the White House can't block incoming calls from a specific number.
(Though even that's not really a great analogy, because phone numbers are not considered to correspond uniquely or permanently to an individual the way that social media accounts are).