[llvm] [AArch64][PAC] Move emission of LR checks in tail calls to AsmPrinter (PR #110705)

[Anatoly Trosinenko via llvm-commits]

================
@@ -169,13 +169,13 @@ define i32 @test_tailcall_ib_var(ptr %arg0, ptr %arg1) #0 {
 
 define void @test_tailcall_omit_mov_x16_x16(ptr %objptr) #0 {
 ; CHECK-LABEL: test_tailcall_omit_mov_x16_x16:
-; CHECK:         ldr     x16, [x0]
-; CHECK:         mov     x17, x0
-; CHECK:         movk    x17, #6503, lsl #48
-; CHECK:         autda   x16, x17
-; CHECK:         ldr     x2, [x16]
-; CHECK:         movk    x16, #54167, lsl #48
-; CHECK:         braa    x2, x16
+; CHECK-NEXT:    ldr     x16, [x0]
----------------
atrosinenko wrote:

This particular test case was affected by this PR, namely at the commit 0268b5be32f the code was generated as
```
test_tailcall_omit_mov_x16_x16:
        ldr     x16, [x0]
        mov     x17, x0
        movk    x17, #6503, lsl #48
        autda   x16, x17
        ldr     x2, [x16]
        mov     x1, x16
        mov     x16, x1
        movk    x16, #54167, lsl #48
        braa    x2, x16
```
Meaning it became even worse that just `mov x16, x16` and still wasn't caught by the `CHECK:` lines. That was one of the reasons to evaluate whether restricting the call target operand works better.

https://github.com/llvm/llvm-project/pull/110705