Spacechains do burn bitcoins but I don't think they are an attack on bitcoin. I also don't recommend burning bitcoins on mainnet, because you'll probably never get back everything you burned. It's fine on testnet though.

Lack of self control. When I originally heard about spacechains I was like, "Burn bitcoins? That's a dumb idea," and then I started thinking up reasons why it's dumb, and then I wondered, "How would you even do that?" so I considered it a few minutes and realized how it might be done, and then I was done for: I had to try it and see if it would work. And it does work, but that doesn't make it a good idea.

That's fine, you can oppose it. I think a few things are worth pointing out in its defense:

(1) This is only on testnet. It's meant as a technical demonstration of one way to do permissionless sidechains without a soft fork. I don't want people to actually burn real bitcoins, as in, on mainnet. But the testnet is precisely for experiments like this. I have no problem with burning testnet bitcoins because stuff like that is precisely what they exist for.

(2) Even if someone does this on mainnet, what will probably happen is this: spacechain creators will do it to issue an initial supply of coins on their spacechain, but after that, there's no reason anyone else has to do it. Other people can just buy some of the intial supply and use those. The option to make your own spacecoins (through burning) is just there to prevent the people who hold the initial supply from charging too much for them. If they ever charge more than 1 sat per spacecoin, you can just make your own spacecoins. Having that option keeps the whole thing permissionless.

(3) You can get rid of the bitcoin burning part by issuing some other token in a trusted manner. For example, tether corp could create a spacechain whose only token is tethers, and they wouldn't have to burn any bitcoins to do that. But this idea is probably even worse than burning bitcoins because it means literally creating an altcoin. (BTW the stacks blockchain is basically a spacechain with their own premined token, so this idea of "issue an altcoin on a spacechain so you don't have to burn any bitcoins" isn't even new. It's just worse than anything anyone ever invented before.)

(4) This is actually the same thing as #3 but one potentially-interesting type of token you could issue on a spacechain is a federated bitcoin-pegged token, similar to liquid bitcoin or rootstock bitcoin. This might be cool to do because a lot of bitcoiners, including myself, don't trust the federations that control liquid and rootstock. But what if you made your own liquid or your own rootstock where you are part of the federation in control? If you do that, you're placing more trust in yourself, which seems potentially appealing. At some point I want to add an option to the spacechain launcher to create a federated peg in this manner.

(5) The reason why I personally think burning bitcoins is dumb is because I don't see any meaningful value I can get in compensation for it. If a spacechain really offered some feature that I thought was worth $10, I would be willing to burn up to $10 in bitcoin to get that feature, because that just seems like common sense to me. The exact method I use to purchase something doesn't matter to me that much, whether it's sending bitcoins to a burn address or to someone who is selling the thing I want. So if someone comes up with a really compelling use case for spacechains (which I doubt will happen) I might burn a few sats to use it, if that's the cheapest way to use the feature. So I don't think burning bitcoins is inherently dumb, I just doubt anyone will come up with anything useful enough to justify it according to my personal preferences.

What problems does a sidechain solve?

This is really interesting.