[https://nvbugs/5769712][fix] fix timeout in AutoDeploy llama accuracy test

Signed-off-by: Lucas Liebenwein <11156568+lucaslie@users.noreply.github.com>

Author: lucaslie

tests/integration/defs/accuracy/test_llm_api_autodeploy.py

@@ -44,7 +44,7 @@ def get_default_kwargs(self, enable_chunked_prefill=False):
                 },
                 "compile_model": {
                     "backend":
-                    "torch-opt",
+                    "torch-cudagraph",
                     "cuda_graph_batch_sizes":
                     [1, 2, 4, 8, 16, 32, 64, 128, 256, 512],
                 },

tests/integration/test_lists/waives.txt

@@ -227,7 +227,6 @@ full:H100_PCIe/unittest/llmapi/test_llm_pytorch.py::test_llama_7b_multi_lora_evi
 unittest/_torch/speculative/test_draft_len_schedule.py::test_correctness_across_batch_sizes[model_drafter-schedule1] SKIP (https://nvbugs/5680911)
 accuracy/test_llm_api_pytorch.py::TestSeedOss_36B::test_auto_dtype SKIP (https://nvbugs/5612438)
 accuracy/test_llm_api_autodeploy.py::TestNemotronH::test_auto_dtype[True] SKIP (https://nvbugs/5688721)
-accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_auto_dtype[False-4] SKIP (https://nvbugs/5769712)
 test_e2e.py::test_openai_completions_example[trt] SKIP (https://nvbugs/5701450)
 accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_4gpus[tp4-fp8kv=False-attn_backend=TRTLLM-torch_compile=False] SKIP (https://nvbugs/5701457)
 triton_server/test_triton_llm.py::test_llmapi_backend[4-0-disableDecoupleMode-tensorrt_llm] SKIP (https://nvbugs/5701480)