Use CommentRanges in backwards lexing (#7360)

## Summary

The tokenizer was split into a forward and a backwards tokenizer. The
backwards tokenizer uses the same names as the forwards ones (e.g.
`next_token`). The backwards tokenizer gets the comment ranges that we
already built to skip comments.

---------

Co-authored-by: Micha Reiser <micha@reiser.io>

crates/ruff_python_formatter/src/comments/debug.rs

@@ -182,6 +182,7 @@ mod tests {
     use ruff_formatter::SourceCode;
     use ruff_python_ast::node::AnyNode;
     use ruff_python_ast::{StmtBreak, StmtContinue};
+    use ruff_python_trivia::CommentRanges;
     use ruff_text_size::{TextRange, TextSize};
 
     use crate::comments::map::MultiMap;
@@ -231,7 +232,8 @@ break;
             ),
         );
 
-        let comments = Comments::new(comments_map);
+        let comment_ranges = CommentRanges::default();
+        let comments = Comments::new(comments_map, &comment_ranges);
 
         assert_debug_snapshot!(comments.debug(source_code));
     }