liblzma: Use 8-byte method in memcmplen.h on ARM64.

It requires fast unaligned access to 64-bit integers
and a fast instruction to count leading zeros in
a 64-bit integer (__builtin_ctzll()). This perhaps
should be enabled on some other archs too.

Thanks to Chenxi Mao for the original patch:
#75 (the first commit)
According to the numbers there, this may improve encoding
speed by about 3-5 %.

This enables the 8-byte method on MSVC ARM64 too which
should work but wasn't tested.

src/liblzma/common/memcmplen.h

@@ -24,7 +24,8 @@
 // can use the intrinsics without the header file.
 #if defined(TUKLIB_FAST_UNALIGNED_ACCESS) \
 		&& defined(_MSC_VER) \
-		&& defined(_M_X64) \
+		&& (defined(_M_X64) \
+			|| defined(_M_ARM64) || defined(_M_ARM64EC)) \
 		&& !defined(__INTEL_COMPILER)
 #	include <intrin.h>
 #endif
@@ -58,20 +59,21 @@ lzma_memcmplen(const uint8_t *buf1, const uint8_t *buf2,
 
 #if defined(TUKLIB_FAST_UNALIGNED_ACCESS) \
 		&& (((TUKLIB_GNUC_REQ(3, 4) || defined(__clang__)) \
-				&& defined(__x86_64__)) \
+				&& (defined(__x86_64__) \
+					|| defined(__aarch64__))) \
 			|| (defined(__INTEL_COMPILER) && defined(__x86_64__)) \
 			|| (defined(__INTEL_COMPILER) && defined(_M_X64)) \
-			|| (defined(_MSC_VER) && defined(_M_X64)))
-	// I keep this x86-64 only for now since that's where I know this
-	// to be a good method. This may be fine on other 64-bit CPUs too.
-	// On big endian one should use xor instead of subtraction and switch
-	// to __builtin_clzll().
+			|| (defined(_MSC_VER) && (defined(_M_X64) \
+				|| defined(_M_ARM64) || defined(_M_ARM64EC))))
+	// This is only for x86-64 and ARM64 for now. This might be fine on
+	// other 64-bit processors too. On big endian one should use xor
+	// instead of subtraction and switch to __builtin_clzll().
 #define LZMA_MEMCMPLEN_EXTRA 8
 	while (len < limit) {
 		const uint64_t x = read64ne(buf1 + len) - read64ne(buf2 + len);
 		if (x != 0) {
 	// MSVC or Intel C compiler on Windows
-#	if (defined(_MSC_VER) || defined(__INTEL_COMPILER)) && defined(_M_X64)
+#	if defined(_MSC_VER) || defined(__INTEL_COMPILER)
 			unsigned long tmp;
 			_BitScanForward64(&tmp, x);
 			len += (uint32_t)tmp >> 3;