Syntax for benchmarking multiple commands

Hey there. Firstly, thanks for the great utility.

I'm confused by the syntax for benchmarking multiple commands, or rather the logic behind it, and the documentation is minimal doesn't explain any further. As far as I can tell, when benchmarking several commands, exactly one of the commands must be enclosed in quotes, and the others cannot be. This can be a little confusing logically, so I was wondering what the reason is for why it's done this way, and whether there's any difference in how benchmarking is done on the command in quotes compared to the other commands? Are there particular commands that should or should not be in quotes?

Thanks in advance, I'd really appreciate it.

Start by reading the existing command-line syntax documentation and the entry point that parses multiple benchmark commands. Verify how quoted and unquoted commands are handled and whether quoting changes benchmarking behavior; document the reason, supported forms, and examples so users can tell which commands require quotes.

Written by the indexing model from the issue text.