[V1] Adding min tokens/repetition/presence/frequence penalties to V1 sampler (#10681)

Signed-off-by: Sourashis Roy <sroy@roblox.com>
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>

vllm/v1/sample/metadata.py

@@ -1,5 +1,5 @@
 from dataclasses import dataclass
-from typing import Dict
+from typing import Dict, List, Optional, Set
 
 import torch
 
@@ -19,3 +19,13 @@ class SamplingMetadata:
     generators: Dict[int, torch.Generator]
 
     max_num_logprobs: int
+
+    no_penalties: bool
+    prompt_token_ids: Optional[torch.Tensor]
+    frequency_penalties: torch.Tensor
+    presence_penalties: torch.Tensor
+    repetition_penalties: torch.Tensor
+
+    output_token_ids: List[List[int]]
+    min_tokens: List[int]
+    stop_token_ids: List[Set[int]]