[llvm] [NVPTX] Add sm_88 and sm_110* arch supports (PR #156857)

[Artem Belevich via llvm-commits]

================
@@ -127,6 +127,7 @@ def : Proc<"sm_75",   [SM75, PTX63]>;
 def : Proc<"sm_80",   [SM80, PTX70]>;
 def : Proc<"sm_86",   [SM86, PTX71]>;
 def : Proc<"sm_87",   [SM87, PTX74]>;
+def : Proc<"sm_88",   [SM88, PTX90]>;
----------------
Artem-B wrote:

That's what it says **now**

Compare to what it said when sm_101 was introduced:
https://docs.nvidia.com/cuda/archive/12.8.0/parallel-thread-execution/index.html
<img width="1964" height="1148" alt="image" src="https://github.com/user-attachments/assets/56fb10d8-4e75-494d-a1e7-87b204ea8c2f" />



https://github.com/llvm/llvm-project/pull/156857