[CIR][Lowering][Bugfix] Fix lowering of bool_to_int cast (llvm#450)

The minimal bug repro:
```
#include <stdbool.h>
#include <stdint.h>
void bar() {
  bool x = true;
  uint8_t y = (uint8_t)x;
}
```
Fails on verification stage:
```
loc("repro.c":5:24): error: integer width of the output type is smaller or equal to the integer width of the input type
fatal error: error in backend: The pass manager failed to lower CIR to LLVMIR dialect!
```
The problem is that in some cases lowering from CIR emits the invalid
zext operation. PR fixes this issue by emitting the llvm.bitcast instead
of llvm.zext in such cases.

clang/lib/CIR/Lowering/DirectToLLVM/LowerToLLVM.cpp

@@ -664,9 +664,15 @@ class CIRCastOpLowering : public mlir::OpConversionPattern<mlir::cir::CastOp> {
     case mlir::cir::CastKind::bool_to_int: {
       auto dstTy = castOp.getType().cast<mlir::cir::IntType>();
       auto llvmSrcVal = adaptor.getOperands().front();
-      auto llvmDstTy = getTypeConverter()->convertType(dstTy);
-      rewriter.replaceOpWithNewOp<mlir::LLVM::ZExtOp>(castOp, llvmDstTy,
-                                                      llvmSrcVal);
+      auto llvmSrcTy = llvmSrcVal.getType().cast<mlir::IntegerType>();
+      auto llvmDstTy =
+          getTypeConverter()->convertType(dstTy).cast<mlir::IntegerType>();
+      if (llvmSrcTy.getWidth() == llvmDstTy.getWidth())
+        rewriter.replaceOpWithNewOp<mlir::LLVM::BitcastOp>(castOp, llvmDstTy,
+                                                           llvmSrcVal);
+      else
+        rewriter.replaceOpWithNewOp<mlir::LLVM::ZExtOp>(castOp, llvmDstTy,
+                                                        llvmSrcVal);
       return mlir::success();
     }
     case mlir::cir::CastKind::bool_to_float: {

clang/test/CIR/Lowering/cast.cir

@@ -81,4 +81,19 @@ module {
     %19 = cir.load %2 : cir.ptr <!s32i>, !s32i
     cir.return %19 : !s32i
   }
+
+  cir.func @testBoolToIntCast(%arg0: !cir.bool)  {
+  // CHECK: llvm.func @testBoolToIntCast
+    %0 = cir.alloca !cir.bool, cir.ptr <!cir.bool>, ["bl", init] {alignment = 1 : i64}
+    %1 = cir.alloca !u8i, cir.ptr <!u8i>, ["y", init] {alignment = 1 : i64}
+    cir.store %arg0, %0 : !cir.bool, cir.ptr <!cir.bool>
+
+    %2 = cir.load %0 : cir.ptr <!cir.bool>, !cir.bool
+    %3 = cir.cast(bool_to_int, %2 : !cir.bool), !u8i
+    // CHECK: %[[LOAD_BOOL:.*]] = llvm.load %{{.*}} : !llvm.ptr -> i8
+    // CHECK: %{{.*}} = llvm.bitcast %[[LOAD_BOOL]] : i8 to i8
+
+    cir.store %3, %1 : !u8i, cir.ptr <!u8i>
+    cir.return
+  }
 }