[llvm-mca][AArch64] Refactor Neoverse tests to split out common inputs (NFC)

For many of these tests the inputs are very similiar with slight
divergences, we should work to make common sources where we can to avoid
subtle differences.

Changes:

- For basic tests there's a relatively large diff between V1 and all
  other cores because of 24f0901 (llvm#128892) which makes it more complete.
  I've been thru the entire diff and 99% of the time V1 makes more
  sense, except for a couple of small changes (might post a separate
  patch for). Therefore I decided it's best to take
  V1-basic-instructions.s as the common source.
- Split out FEAT_RCPC_IMMO tests from basic since N1 doesnt have this feature.
- Split out FEAT_MTE tests. V2/V3 also have this feature but were
  missing tests, so I've added them.

The NEON and SVE tests are also quite substantial and could be made
common, but there's also subtle differences that take time to go thru.

Author: c-rhodes

llvm/test/tools/llvm-mca/AArch64/Neoverse/Inputs/basic-instructions.s

@@ -0,0 +1,105 @@
+irg	x0, x1
+irg	sp, x1
+irg	x0, sp
+irg	x0, x1, x2
+irg	sp, x1, x2
+addg	x0, x1, #0, #1
+addg	sp, x2, #32, #3
+addg	x0, sp, #64, #5
+addg	x3, x4, #1008, #6
+addg	x5, x6, #112, #15
+subg	x0, x1, #0, #1
+subg	sp, x2, #32, #3
+subg	x0, sp, #64, #5
+subg	x3, x4, #1008, #6
+subg	x5, x6, #112, #15
+gmi	x0, x1, x2
+gmi	x3, sp, x4
+gmi	xzr, x0, x30
+gmi	x30, x0, xzr
+subp	x0, x1, x2
+subps	x0, x1, x2
+subp	x0, sp, sp
+subps	x0, sp, sp
+subps	xzr, x0, x1
+subps	xzr, sp, sp
+stg	x0, [x1, #-4096]
+stg	x1, [x2, #4080]
+stg	x2, [sp, #16]
+stg	x3, [x1]
+stg	sp, [x1]
+stzg	x0, [x1, #-4096]
+stzg	x1, [x2, #4080]
+stzg	x2, [sp, #16]
+stzg	x3, [x1]
+stzg	sp, [x1]
+stg	x0, [x1, #-4096]!
+stg	x1, [x2, #4080]!
+stg	x2, [sp, #16]!
+stg	sp, [sp, #16]!
+stzg	x0, [x1, #-4096]!
+stzg	x1, [x2, #4080]!
+stzg	x2, [sp, #16]!
+stzg	sp, [sp, #16]!
+stg	x0, [x1], #-4096
+stg	x1, [x2], #4080
+stg	x2, [sp], #16
+stg	sp, [sp], #16
+stzg	x0, [x1], #-4096
+stzg	x1, [x2], #4080
+stzg	x2, [sp], #16
+stzg	sp, [sp], #16
+st2g	x0, [x1, #-4096]
+st2g	x1, [x2, #4080]
+st2g	x2, [sp, #16]
+st2g	x3, [x1]
+st2g	sp, [x1]
+stz2g	x0, [x1, #-4096]
+stz2g	x1, [x2, #4080]
+stz2g	x2, [sp, #16]
+stz2g	x3, [x1]
+stz2g	sp, [x1]
+st2g	x0, [x1, #-4096]!
+st2g	x1, [x2, #4080]!
+st2g	x2, [sp, #16]!
+st2g	sp, [sp, #16]!
+stz2g	x0, [x1, #-4096]!
+stz2g	x1, [x2, #4080]!
+stz2g	x2, [sp, #16]!
+stz2g	sp, [sp, #16]!
+st2g	x0, [x1], #-4096
+st2g	x1, [x2], #4080
+st2g	x2, [sp], #16
+st2g	sp, [sp], #16
+stz2g	x0, [x1], #-4096
+stz2g	x1, [x2], #4080
+stz2g	x2, [sp], #16
+stz2g	sp, [sp], #16
+stgp	x0, x1, [x2, #-1024]
+stgp	x0, x1, [x2, #1008]
+stgp	x0, x1, [sp, #16]
+stgp	xzr, x1, [x2, #16]
+stgp	x0, xzr, [x2, #16]
+stgp	x0, xzr, [x2]
+stgp	x0, x1, [x2, #-1024]!
+stgp	x0, x1, [x2, #1008]!
+stgp	x0, x1, [sp, #16]!
+stgp	xzr, x1, [x2, #16]!
+stgp	x0, xzr, [x2, #16]!
+stgp	x0, x1, [x2], #-1024
+stgp	x0, x1, [x2], #1008
+stgp	x0, x1, [sp], #16
+stgp	xzr, x1, [x2], #16
+stgp	x0, xzr, [x2], #16
+ldg	x0, [x1]
+ldg	x2, [sp, #-4096]
+ldg	x3, [x4, #4080]
+ldgm	x0, [x1]
+ldgm	x1, [sp]
+ldgm	xzr, [x2]
+stgm	x0, [x1]
+stgm	x1, [sp]
+stgm	xzr, [x2]
+stzgm	x0, [x1]
+stzgm	x1, [sp]
+stzgm	xzr, [x2]

llvm/test/tools/llvm-mca/AArch64/Neoverse/Inputs/mte-instructions.s

@@ -0,0 +1,17 @@
+#------------------------------------------------------------------------------
+# Load/store exclusive
+#------------------------------------------------------------------------------
+
+ldapur     w7, [x24]
+ldapur     x20, [x13]
+ldapurb    w13, [x17]
+ldapurh    w3, [x22]
+ldapursb   w7, [x8]
+ldapursb   x29, [x7]
+ldapursh   w17, [x19]
+ldapursh   x3, [x3]
+ldapursw   x3, [x18]
+stlur      w3, [x27]
+stlur      x23, [x25]
+stlurb     w30, [x17]
+stlurh     w9, [x29]