make the shard sha256 extract endianness-agnostig

We should have interpreted it as big endian right from the start,
but now that we got the existing code developed on little endian,
we keep it so.

noticed after varnish core merge

src/shard_hash.c

@@ -42,6 +42,17 @@
 #include "parse_vcc_enums.h"
 #include "shard_hash.h"
 
+static __inline void
+vle32enc(void *pp, uint32_t u)
+{
+	uint8_t *p = (uint8_t *)pp;
+
+	p[0] = u & 0xff;
+	p[1] = (u >> 8) & 0xff;
+	p[2] = (u >> 16) & 0xff;
+	p[3] = (u >> 24) & 0xff;
+}
+
 static uint32_t __match_proto__(hash_func)
 shard_hash_crc32(VCL_STRING s)
 {
@@ -61,6 +72,7 @@ shard_hash_sha256(VCL_STRING s)
 		unsigned char digest[32];
 		uint32_t uint32_digest[8];
 	} sha256_digest;
+	uint32_t r;
 
 	SHA256_Init(&sha256);
 	SHA256_Update(&sha256, s, strlen(s));
@@ -70,7 +82,8 @@ shard_hash_sha256(VCL_STRING s)
 	 * use low 32 bits only
 	 * XXX: Are these the best bits to pick?
 	 */
-	return (sha256_digest.uint32_digest[7]);
+	vle32enc(&r, sha256_digest.uint32_digest[7]);
+	return (r);
 }
 
 static uint32_t __match_proto__(hash_func)