Basic SIMD reduction support.

Rationale:
Enables vectorization of x += .... for very basic (simple, same-type)
constructs. Paves the way for more complex (narrower and/or mixed-type)
constructs, which will be handled by the next CL.

This is  a revert   of Icb5d6c805516db0a1d911c3ede9a246ccef89a22
and thus a revert^2 of I2454778dd0ef1da915c178c7274e1cf33e271d0f
and thus a revert^3 of I1c1c87b6323e01442e8fbd94869ddc9e760ea1fc
and thus a revert^4 of I7880c135aee3ed0a39da9ae5b468cbf80e613766

PS1-2 shows what needed to change

Test: test-art-host test-art-target

Bug: 64091002
Change-Id: I647889e0da0959ca405b70081b79c7d3c9bcb2e9
diff --git a/compiler/optimizing/scheduler_arm64.cc b/compiler/optimizing/scheduler_arm64.cc
index 510619f..1d9d28a 100644
--- a/compiler/optimizing/scheduler_arm64.cc
+++ b/compiler/optimizing/scheduler_arm64.cc
@@ -215,12 +215,12 @@
   last_visited_latency_ = kArm64SIMDReplicateOpLatency;
 }
 
-void SchedulingLatencyVisitorARM64::VisitVecSetScalars(HVecSetScalars* instr) {
-  LOG(FATAL) << "Unsupported SIMD instruction " << instr->GetId();
+void SchedulingLatencyVisitorARM64::VisitVecExtractScalar(HVecExtractScalar* instr) {
+  HandleSimpleArithmeticSIMD(instr);
 }
 
-void SchedulingLatencyVisitorARM64::VisitVecSumReduce(HVecSumReduce* instr) {
-  LOG(FATAL) << "Unsupported SIMD instruction " << instr->GetId();
+void SchedulingLatencyVisitorARM64::VisitVecReduce(HVecReduce* instr) {
+  HandleSimpleArithmeticSIMD(instr);
 }
 
 void SchedulingLatencyVisitorARM64::VisitVecCnv(HVecCnv* instr ATTRIBUTE_UNUSED) {
@@ -283,8 +283,8 @@
   last_visited_latency_ = kArm64SIMDIntegerOpLatency;
 }
 
-void SchedulingLatencyVisitorARM64::VisitVecAndNot(HVecAndNot* instr) {
-  LOG(FATAL) << "Unsupported SIMD instruction " << instr->GetId();
+void SchedulingLatencyVisitorARM64::VisitVecAndNot(HVecAndNot* instr ATTRIBUTE_UNUSED) {
+  last_visited_latency_ = kArm64SIMDIntegerOpLatency;
 }
 
 void SchedulingLatencyVisitorARM64::VisitVecOr(HVecOr* instr ATTRIBUTE_UNUSED) {
@@ -307,6 +307,10 @@
   HandleSimpleArithmeticSIMD(instr);
 }
 
+void SchedulingLatencyVisitorARM64::VisitVecSetScalars(HVecSetScalars* instr) {
+  HandleSimpleArithmeticSIMD(instr);
+}
+
 void SchedulingLatencyVisitorARM64::VisitVecMultiplyAccumulate(
     HVecMultiplyAccumulate* instr ATTRIBUTE_UNUSED) {
   last_visited_latency_ = kArm64SIMDMulIntegerLatency;