[llvm] [llvm] Ensure that soft float targets don't use float/vector code for memops. (PR #107022)

[Matt Arsenault via llvm-commits]

================
@@ -2023,6 +2023,11 @@ class TargetLoweringBase {
     return LLT();
   }
 
+  bool useIntScalarMemOps(const AttributeList &FuncAttributes) const {
----------------
arsenm wrote:

useSoftFloat is better but also vague. use "soft" float for what? 

https://github.com/llvm/llvm-project/pull/107022