runtime: faster memclr on x86.

Use explicit SSE writes instead of REP STOSQ.

benchmark               old ns/op    new ns/op    delta
BenchmarkMemclr5               22            5  -73.62%
BenchmarkMemclr16              27            5  -78.49%
BenchmarkMemclr64              28            6  -76.43%
BenchmarkMemclr256             34            8  -74.94%
BenchmarkMemclr4096           112           84  -24.73%
BenchmarkMemclr65536         1902         1920   +0.95%

LGTM=dvyukov
R=golang-codereviews, dvyukov
CC=golang-codereviews
https://golang.org/cl/60090044

src/pkg/runtime/export_test.go

@@ -84,3 +84,7 @@ func GogoBytes() int32
 
 var hashLoad float64 // declared in hashmap.c
 var HashLoad = &hashLoad
+
+func memclrBytes(b []byte)
+
+var MemclrBytes = memclrBytes