First of all, @sbdpotato and @burnpost are quite different, and I'm not sure they should even be lumped together in any discussion. @sbdpotato is raising funds, and is now getting most of its funds from SPS, which makes capping payouts relatively harmless. By contrast, @burnpost raises no funds, and is instead a method by which voters can vote for 'None of the above' when it comes to distributing rewards. Capping rewards would be overriding the will of the voters and IMO @burnpost should rather remain neutral. If you think @burnpost is getting too many votes, take it up with the voters, or create better, more compelling content that is more attractive to voters than 'None of the above'. (I've already stated my view that probably the best solution overall is for "Trending" to be renamed "Most voted" or "Most rewards", which would be a more accurate description of what it actually means, and get rid of the misleading implication that it is actually the most interesting or attention-worthy content.)
I added a line suggesting that voters consider later votes to reduce trending. The burnposts are now down to one per day, along with comments (but the comments tend to trend less). To me that seems like pretty reasonable accomodations to those who want to minimize the impact on Trending, but I'm open to other constructive suggestions.
On your claim that 'the argument against is mounting' I think we will have to agree to disagree. From my perspective, the true measure of support is the voting, and everything else is a bunch of hot air. As long as it continues to get votes, that tells me there is significant (even in some sense very large) support. As you can see from Chart 9 here, not only does it continue to get votes, but support for burning has generally continued to increase.
As far as getting serious about on-boarding, I think that is nonsense. There are only a small number of accounts created per day and a good chunk of those are created by @steemmonsters, which doesn't even use the social rewards system at all (data is here). And of those small number of accounts created, very few ever become or stay active users.
In four years we have had all sorts of different populations of posts on Trending, none of which has ever done much to drive growth or increase successful onboarding, and after four years I doubt very much that we are going to ever see Trending be a true driver of growth. Without a new UI and/or use case for Steem becoming popular (which could be revised steemit.com, could be another site, could be games such as steemmonsters), I don't see this changing. Voters seem to agree, and increasingly prefer to preserve some value in the Steem ecosystem by reducing selling pressure and burning rather than blindly continue a failed model.