Implement autoformatter for tree-sitter query language

[pokey]

It's a bit cumbersome to write our new tree-sitter queries (#616) because there is no autoformatter. I don't know of one today, but we could probably write one by leveraging the tree-sitter parser for tree-sitter-query. We might want to build the formatter using dprint. A couple possibly useful resources on dprint:

• https://github.com/dprint/dprint/blob/main/docs/overview.md
• https://github.com/dprint/dprint/blob/main/docs/wasm-plugin-development.md
• https://docs.rs/dprint-core/0.28.0/dprint_core/formatting/index.html
• VSCode support

Alternately, we might look into using the machinery that @wenkokke developed for https://github.com/wenkokke/talonfmt

This autoformatter would obviously be useful outside of Cursorless for the many other projects writing tree-sitter queries

Update: today we're relying on @AndreasArvidsson's VSCode extension to do autoformatting for scm files, which is great for development, but doesn't support CI.

Also, per @auscompgeek's suggestion, we may want to consider writing the formatter as a prettier plugin. I poked through the docs a bit and from what I saw it looks like it's potentially a nice fit

We may also consider using https://topiary.tweag.io/, which has experimental support for formatting .scm files, but as of 2023-12-01 doesn't yet have a VSCode extension (see tweag/topiary#655 to track)