If you're an engineer for Google or Apple the best thing you could do is build into an easy way for users to report to the FCC. It's a routine forum, no ML needed but hey, pitch it if you need that to get this shit done.
If you work somewhere that is spamming and enshitified, the way to convince your boss is to show them that their domains are being blocked and that leads to less money. That's the language they understand. They don't understand metrics (that's how we got here in the first place. So don't get technical!)
Unfortunately if you make reporting too easy whoever processes the reports will have to deal with a whole lot of false positives. A lot of people seem to confuse spam and non-spam that they just happen to not be interested in at the moment.
For example if you sell things on the web people will come to your site, add items to your cart, go to your checkout page, enter their email, pay, and then when they receive an emailed receipt mark it spam.
Heck, I've seen people who ask for help by sending email to a support email address, and then mark the automatic reply from the ticketing system that lets them know their request has been received and tells them how long it will take before someone gets back to them as spam.
I've also seen people complain that a company wasn't responding to their emails to support, when in fact the company is responding but the person had marked earlier receipt emails or emails letting them no that their subscription that they had opted into auto-renew for was going to renew soon as spam.
Note: in all the examples above I'm talking about companies that do not send marketing emails except to people who go to a separate page that is explicitly for signing up for newsletters. The only emails sent to the person where receipts, re-bill notices, and responses to mails the person had sent.