Move string-prefix enumerations to a separate submodule (#11425)

## Summary

This moves the string-prefix enumerations in `ruff_python_ast` to a
separate submodule. I think this helps clarify that these prefixes are
purely abstract: they only depend on each other, and do not depend on
any of the other code in `nodes.rs` in any way. Moreover, while various
AST nodes _use_ them, they're not really nodes themselves, so they feel
slightly out of place in `nodes.rs`.

I considered moving all of them to `str.rs`, but it felt like enough
code that it could be a separate submodule.

## Test Plan

`cargo test`

crates/ruff_python_formatter/src/string/mod.rs

@@ -2,7 +2,11 @@ pub(crate) use any::AnyString;
 pub(crate) use normalize::{normalize_string, NormalizedString, StringNormalizer};
 use ruff_formatter::format_args;
 use ruff_python_ast::str::Quote;
-use ruff_python_ast::{self as ast, AnyStringFlags, AnyStringPrefix};
+use ruff_python_ast::{
+    self as ast,
+    str_prefix::{AnyStringPrefix, StringLiteralPrefix},
+    AnyStringFlags,
+};
 use ruff_text_size::{Ranged, TextRange};
 
 use crate::comments::{leading_comments, trailing_comments};
@@ -58,9 +62,7 @@ impl Format<PyFormatContext<'_>> for AnyStringPrefix {
         // Remove the unicode prefix `u` if any because it is meaningless in Python 3+.
         if !matches!(
             self,
-            AnyStringPrefix::Regular(
-                ast::StringLiteralPrefix::Empty | ast::StringLiteralPrefix::Unicode
-            )
+            AnyStringPrefix::Regular(StringLiteralPrefix::Empty | StringLiteralPrefix::Unicode)
         ) {
             token(self.as_str()).fmt(f)?;
         }

crates/ruff_python_formatter/src/string/normalize.rs

@@ -693,7 +693,11 @@ impl UnicodeEscape {
 mod tests {
     use std::borrow::Cow;
 
-    use ruff_python_ast::{str::Quote, AnyStringFlags, AnyStringPrefix, ByteStringPrefix};
+    use ruff_python_ast::{
+        str::Quote,
+        str_prefix::{AnyStringPrefix, ByteStringPrefix},
+        AnyStringFlags,
+    };
 
     use super::{normalize_string, UnicodeEscape};