Use ScopedArenaAllocator for register allocation.

Memory needed to compile the two most expensive methods for
aosp_angler-userdebug boot image:
  BatteryStats.dumpCheckinLocked() : 25.1MiB -> 21.1MiB
  BatteryStats.dumpLocked(): 49.6MiB -> 42.0MiB
This is because all the memory previously used by Scheduler
is reused by the register allocator; the register allocator
has a higher peak usage of the ArenaStack.

And continue the "arena"->"allocator" renaming.

Test: m test-art-host-gtest
Test: testrunner.py --host
Bug: 64312607
Change-Id: Idfd79a9901552b5147ec0bf591cb38120de86b01
diff --git a/compiler/optimizing/linearize_test.cc b/compiler/optimizing/linearize_test.cc
index e82fab9..b2a9c0a 100644
--- a/compiler/optimizing/linearize_test.cc
+++ b/compiler/optimizing/linearize_test.cc
@@ -45,7 +45,7 @@
   std::unique_ptr<const X86InstructionSetFeatures> features_x86(
       X86InstructionSetFeatures::FromCppDefines());
   x86::CodeGeneratorX86 codegen(graph, *features_x86.get(), CompilerOptions());
-  SsaLivenessAnalysis liveness(graph, &codegen);
+  SsaLivenessAnalysis liveness(graph, &codegen, GetScopedAllocator());
   liveness.Analyze();
 
   ASSERT_EQ(graph->GetLinearOrder().size(), number_of_blocks);