Use kwargs for long lists of `EngineCoreRequest` arguments in tests and fix extra kwargs (#24987)

Signed-off-by: Andrew Sansom <andrew@protopia.ai>

tests/detokenizer/test_min_tokens.py

@@ -31,16 +31,14 @@ def test_min_tokens_with_stop(min_tokens: int, stop: str, truth: str):
         stop=stop,
         min_tokens=min_tokens,
     )
-    request = EngineCoreRequest("",
-                                prompt_token_ids,
-                                None,
-                                None,
-                                None,
-                                params,
-                                None,
-                                None,
-                                0.0,
-                                None,
+    request = EngineCoreRequest(request_id="",
+                                prompt_token_ids=prompt_token_ids,
+                                mm_features=None,
+                                sampling_params=params,
+                                pooling_params=None,
+                                eos_token_id=None,
+                                arrival_time=0.0,
+                                lora_request=None,
                                 cache_salt=None,
                                 data_parallel_rank=None)
 

tests/tokenization/test_detokenize.py

@@ -61,14 +61,14 @@ def _run_incremental_decode(tokenizer,
         skip_special_tokens=skip_special_tokens,
         spaces_between_special_tokens=spaces_between_special_tokens,
     )
-    request = EngineCoreRequest("",
-                                prompt_token_ids,
-                                None,
-                                params,
-                                None,
-                                None,
-                                0.0,
-                                None,
+    request = EngineCoreRequest(request_id="",
+                                prompt_token_ids=prompt_token_ids,
+                                mm_features=None,
+                                sampling_params=params,
+                                pooling_params=None,
+                                eos_token_id=None,
+                                arrival_time=0.0,
+                                lora_request=None,
                                 cache_salt=None,
                                 data_parallel_rank=None)