"Company declines to do business with me" isn't usually a cause of action you can sue over.
The US legal system is all based on proving harm.
In this case:
1: SSLMate is a paying customer. (Correct me if I'm wrong.)
2: Google harmed SSLMate, and their customers, by deliberately interrupting the services that were paid for.
The big question is if SSLMate was following the terms of service. If SSLMate was actually violating the terms, then it's a hard case to make. Otherwise, Google violated the contract and harmed SSLMate, and is therefore a valid target in US court.
You can sue over anything, doesn't mean you'll win.
If enough people started making companies show up to small claims court for their shitty behaviour maybe they wouldn't act so shitty.
But sue-ing means your problem goes to legal and not tier 1 customer support.
So Mary from legal walks over to Manager whoever and says "Hey why the fuck did you terminate this guys account? Now I have to go to court."
Account reinstated. You drop the case.
This produces stupidly improved results.
https://en.wikipedia.org/wiki/Tortious_interference
covers some situations where someone stops you for making money, for no good reason.
"As an example, someone could ... obstruct someone's ability to honor a contract with a client by deliberately refusing to deliver necessary goods."
For EU citizens, GDPR requires that if you ask for it, a human has to review your case. (Article 22 "The data subject shall have the right not to be subject to a decision based solely on automated processing, including profiling, which produces legal effects concerning him or her or similarly significantly affects him or her.")
I guess a lawyer can argue against this, but I'd say that losing access to a lifetime if mails is absolutely up there with "legal effects concerning him or her or similarly significantly affects him or her."
And from my own experience building software for government services, I can tell you this: In my experience in those systems it is not acceptable to just have a list where someone clicks “deny” all day. Or allow for that matter. We tried with a system were the rule is that the citizen gets <think they apply for> whenever all relevant demands are met. Legal was very clear: No automated decisions either way unless the relevant laws or regulations explicitly allow it, every case has to be reviewed independently — even when the outcome seems completely obvious to anyone who knows the field.
If you reframe this as: a commercial entity progressively dominates a service that is increasingly becoming necessary for survival in the modern world (e.g. primary email / identity provider) or in a given profession (Android developer), and then denies that service to some individuals, while also keeping the cost of switching away to competitors high, then there is a case for natural justice, even if there (still) aren't laws in the books to cover it.