lexer: Tiny improvement to shebang detection

Lexer now discerns between regular comments and doc comments, so use that.
The change only affects the choice of reported errors.

src/test/ui/parser/shebang/shebang-doc-comment.rs

@@ -1,6 +1,3 @@
 #!///bin/bash
 [allow(unused_variables)]
-//~^^ ERROR expected `[`, found doc comment
-
-// Doc comment is misinterpreted as a whitespace (regular comment) during shebang detection.
-// Even if it wasn't, it would still result in an error, just a different one.
+//~^ ERROR expected item, found `[`

src/test/ui/parser/shebang/shebang-doc-comment.stderr

@@ -1,8 +1,8 @@
-error: expected `[`, found doc comment `///bin/bash`
-  --> $DIR/shebang-doc-comment.rs:1:3
+error: expected item, found `[`
+  --> $DIR/shebang-doc-comment.rs:2:1
    |
-LL | #!///bin/bash
-   |   ^^^^^^^^^^^ expected `[`
+LL | [allow(unused_variables)]
+   | ^ expected item
 
 error: aborting due to previous error