negligent refereeing...
claims to have spent 8 months to write a quick opinion to reject
This person may be a troll but the situation is entirely plausible. You have no idea of the volume of quick opinion requests a relatively small number of mathematicians get. (That's your scandal, if you're looking for one. Should c.100 mathematicians be gatekeepers for all the top journals via the quick opinion system? If you're in love with the 'prestige' of journals like Annals etc., that's how it's maintained.) Most mathematicians give priority to quick opinion requests from journals with which they have a formal affiliation, because they've signed up for that job. When a random other journal -- even Annals -- sends a request and doesn't follow up when they don't get a quick reply, that's on them.
What on earth do you mean by "formal affiliation"? You mean like, you're on the editorial board? I'm on the editorial board of a couple of journals (not remotely Annals level!), and they don't ask me for quick opinions, but to handle the paper.
I think the quick opinion system works well. You need a way to quickly reject most papers even from mid-level journals, and that leaves you with two choices other than quick opinions: always get full reports (impossible, people can only write so many referee reports), or just desk reject on the editor's whim.