Merge #8056

8056: completion relevance consider if types can be unified r=JoshMcguigan a=JoshMcguigan

This PR improves completion relevance scoring for generic types, in cases where the types could unify. 

### Before

![pre-could-unify](https://user-images.githubusercontent.com/22216761/111338556-46d94e80-8634-11eb-9936-2b20eb9e6756.png)

### After

![post-could-unify](https://user-images.githubusercontent.com/22216761/111338598-4e005c80-8634-11eb-92e0-69c2c1cda6fc.png)

changelog feature improve completions

Co-authored-by: Josh Mcguigan <[email protected]>

1 files changed, 5 insertions, 1 deletions

diff --git a/crates/hir/src/lib.rs b/crates/hir/src/lib.rs
index 05a60e158..4ee08ef21 100644
--- a/crates/hir/src/lib.rs
+++ b/crates/hir/src/lib.rs
@@ -51,7 +51,7 @@ use hir_def::{
 };
 use hir_expand::{diagnostics::DiagnosticSink, name::name, MacroDefKind};
 use hir_ty::{
-    autoderef,
+    autoderef, could_unify,
     method_resolution::{self, TyFingerprint},
     primitive::UintTy,
     to_assoc_type_id,
@@ -2154,6 +2154,10 @@ impl Type {
 
         walk_type(db, self, &mut cb);
     }
+
+    pub fn could_unify_with(&self, other: &Type) -> bool {
+        could_unify(&self.ty, &other.ty)
+    }
 }
 
 // FIXME: closures