Fix panic in split_imports assist

The fix is admittedly quit literally just papering over.

Long-term, I see two more principled approaches:

* we switch to a fully tree-based impl, without parse . to_string
  step; with this approach, there shouldn't be any panics. The results
  might be nonsensical, but so was the original input.

* we preserve the invariant that re-parsing constructed node is an
  identity, and make all the `make_xxx` method return an `Option`.

closes #4044

crates/ra_syntax/src/ast/edit.rs

@@ -307,7 +307,11 @@ impl ast::UseTree {
 
         fn split_path_prefix(prefix: &ast::Path) -> Option<ast::Path> {
             let parent = prefix.parent_path()?;
-            let mut res = make::path_unqualified(parent.segment()?);
+            let segment = parent.segment()?;
+            if algo::has_errors(segment.syntax()) {
+                return None;
+            }
+            let mut res = make::path_unqualified(segment);
             for p in iter::successors(parent.parent_path(), |it| it.parent_path()) {
                 res = make::path_qualified(res, p.segment()?);
             }