fix(router): retry messages when server is not ready

[imsnif]

This fixes a crash introduced in the latest release where Zellij would sometimes crash on startup with this error:

Error occurred in server:

  × Thread 'server_router' panicked.
  ├─▶ Originating Thread(s)
  │     1. stdin_handler_thread: AcceptInput
  │   
  ├─▶ At zellij-server/src/route.rs:503:30
  ╰─▶ called `Option::unwrap()` on a `None` value
  help: If you are seeing this message, it means that something went wrong.
        Please report this error to the github issue.
        (https://github.com/zellij-org/zellij/issues)
        
        Also, if you want to see the backtrace, you can set the `RUST_BACKTRACE` environment variable to `1`.

This was a race condition that happened when the Zellij client sent a message to the server before it was ready. This was exposed in the latest release because we broke down the queries we make to the terminal emulator (for stuff like pixel-to-cell dimensions and the terminal color palette) into several different messages. This caused them to be sent faster and hit the issue.

This fix involves adding a retry queue for these kinds of messages, placing instructions there when the server is not ready, and retrying them again along with the next messages when they come.

Right now I only implemented this for this one particular message that was causing the issue as an experiment. Care should be taken before duplicating this logic to other instructions.