[SIL] Explicitly use uint64_t for maxBitfieldID

The `maxBitfieldID` was defined as `size_t` and used to assert
bitfield ID values. The `currentBitfieldID` of `SILFunction` is
defined as `uint64_t`, so we should consistently use `uint64_t`
for `maxBitfieldID` as well for 32-bit platforms.

Author: kateinoigakukun

include/swift/SIL/SILBasicBlock.h

@@ -154,7 +154,7 @@ public SwiftObjectHeader {
 
   enum { numCustomBits = std::numeric_limits<CustomBitsType>::digits };
 
-  constexpr static const size_t maxBitfieldID =
+  constexpr static const uint64_t maxBitfieldID =
       std::numeric_limits<uint64_t>::max();
 
   /// Gets the ID (= index in the function's block list) of the block.

include/swift/SIL/SILNode.h

@@ -128,7 +128,7 @@ class alignas(8) SILNode :
 
   enum { numCustomBits = 20 };
 
-  constexpr static const size_t maxBitfieldID =
+  constexpr static const uint64_t maxBitfieldID =
       std::numeric_limits<uint64_t>::max() >> numCustomBits;
 
 protected:

include/swift/SIL/SILValue.h

@@ -1022,7 +1022,7 @@ class Operand {
 public:
   enum { numCustomBits = 8 };
 
-  constexpr static const size_t maxBitfieldID =
+  constexpr static const uint64_t maxBitfieldID =
       std::numeric_limits<uint64_t>::max() >> numCustomBits;
 
 private: