AArch64Subtarget: Check ARMProcFamily instead of CPU part of triple.

[MatzeB]

Summary

A bunch of isXXX() checks in AArch64Subtarget compare the CPU part of
the target triple. It seems like we can use the existing (but currently
unused) ARMProcFamily field instead and save the string comparison.

This is NFC, except for the case of CPU=="cortex-a72" which is reported
as ARMProcFamilty==CortexA57 which lets isCortexA57() report true after
this commit.

[MatzeB]

This is superseded by http://reviews.llvm.org/D20762

lib/Target/AArch64/AArch64Subtarget.h

public:
   bool isTargetWindows() const { return TargetTriple.isOSWindows(); }
   bool isTargetAndroid() const { return TargetTriple.isAndroid(); }
 
   bool isTargetCOFF() const { return TargetTriple.isOSBinFormatCOFF(); }
   bool isTargetELF() const { return TargetTriple.isOSBinFormatELF(); }
   bool isTargetMachO() const { return TargetTriple.isOSBinFormatMachO(); }
 
   bool isGeneric() const { return CPUString == "generic"; }
-  bool isCyclone() const { return CPUString == "cyclone"; }
-  bool isCortexA57() const { return CPUString == "cortex-a57"; }
-  bool isCortexA53() const { return CPUString == "cortex-a53"; }
-  bool isExynosM1() const { return CPUString == "exynos-m1"; }
-  bool isKryo() const { return CPUString == "kryo"; }
+  bool isCyclone() const { return ARMProcFamily == Cyclone; }
+  bool isCortexA57() const { return ARMProcFamily == CortexA57; }
+  bool isCortexA53() const { return ARMProcFamily == CortexA53; }
+  bool isExynosM1() const { return ARMProcFamily == ExynosM1; }
+  bool isKryo() const { return ARMProcFamily == Kryo; }
 
   bool useAA() const override { return isCortexA53(); }
 
   /// getMaxInlineSizeThreshold - Returns the maximum memset / memcpy size
   /// that still makes it profitable to inline the call.
   unsigned getMaxInlineSizeThreshold() const { return 64; }
 
   /// ParseSubtargetFeatures - Parses features string setting specified