[Core][2/N] Model runner refactoring part 2. Combine prepare prefill / decode to a single API (#4681)

This PR combines prepare_prompt and prepare_decode into a single API. This PR also coelsce the attn metadata for prefill/decode to a single class and allow to slice them when running attn backend.

It also refactors subquery_start_loc which was not refactored in the previous PR

vllm/engine/arg_utils.py

@@ -618,6 +618,11 @@ class EngineArgs:
         decoding_config = DecodingConfig(
             guided_decoding_backend=self.guided_decoding_backend)
 
+        if (model_config.get_sliding_window() is not None
+                and scheduler_config.chunked_prefill_enabled):
+            raise ValueError(
+                "Chunked prefill is not supported with sliding window.")
+
         return EngineConfig(model_config=model_config,
                             cache_config=cache_config,
                             parallel_config=parallel_config,