[mlir][sparse] Fixing case coverage warning

Followup to D133835 for fixing the warning on LLVM's Windows buildbot

Reviewed By: aartbik, Peiming, stella.stamenova

Differential Revision: https://reviews.llvm.org/D134925

mlir/include/mlir/ExecutionEngine/SparseTensor/File.h

@@ -44,7 +44,7 @@ namespace sparse_tensor {
 /// as well as providing the buffers and methods for parsing those headers.
 class SparseTensorFile final {
 public:
-  enum class ValueKind {
+  enum class ValueKind : uint8_t {
     // The value before calling `readHeader`.
     kInvalid = 0,
     // Values that can be set by `readMMEHeader`.

mlir/lib/ExecutionEngine/SparseTensor/File.cpp

@@ -107,6 +107,9 @@ bool SparseTensorFile::canReadAs(PrimaryType valTy) const {
     // So we allow implicitly converting the stored values to both
     // integer and floating primary-types.
     return isRealPrimaryType(valTy);
+  default:
+    MLIR_SPARSETENSOR_FATAL("Unknown ValueKind: %d\n",
+                            static_cast<uint8_t>(valueKind_));
   }
 }