Improve endianness check

Update the `hsa.h` header to use the gcc / clang `__BYTE_ORDER__`
macros where available to more accurately autodetect endianness for
the target.

Change-Id: I7312f3badcba9287a30eb14882b91e2a247acc5f


[ROCm/ROCR-Runtime commit: 4971150576]

projects/rocr-runtime/runtime/hsa-runtime/inc/hsa.h

@@ -79,8 +79,12 @@
 
 // Try to detect CPU endianness
 #if !defined(LITTLEENDIAN_CPU) && !defined(BIGENDIAN_CPU)
-#if defined(__i386__) || defined(__x86_64__) || defined(_M_IX86) || \
-    defined(_M_X64) || defined(__loongarch64) || defined(__riscv)
+#if defined(__BYTE_ORDER__) && (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
+#define LITTLEENDIAN_CPU
+#elif defined(__BYTE_ORDER__) && (__BYTE_ORDER__ == __ORDER_BIG_ENDIAN__)
+#define BIGENDIAN_CPU
+#elif defined(__i386__) || defined(__x86_64__) || defined(_M_IX86) || \
+      defined(_M_X64) || defined(__loongarch64) || defined(__riscv)
 #define LITTLEENDIAN_CPU
 #endif
 #endif