[Perf] Use np.ndarray instead of list[list[int]] to reduce GC overhead (#28245)

Signed-off-by: Jialin Ouyang <Jialin.Ouyang@gmail.com>

tests/v1/engine/utils.py

@@ -5,6 +5,7 @@ import random
 from dataclasses import dataclass
 from typing import TypeAlias
 
+import numpy as np
 import torch
 from transformers import PreTrainedTokenizer, PreTrainedTokenizerFast
 
@@ -369,9 +370,9 @@ class MockEngineCore:
                         self.generated_logprobs_raw[req_idx][token_idx]
                     )
                     logprobs = LogprobsLists(
-                        [logprobs_token_ids_],
-                        [logprobs_],
-                        [sampled_token_ranks_],
+                        np.array([logprobs_token_ids_]),
+                        np.array([logprobs_]),
+                        np.array([sampled_token_ranks_]),
                     )
                 else:
                     logprobs = None