[V0 deprecation] Remove long context LoRA (#21169)

Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>

vllm/lora/punica_wrapper/punica_gpu.py

@@ -7,7 +7,7 @@ Punica: Multi-Tenant LoRA Serving.
 https://arxiv.org/abs/2310.18547
 """
 
-from typing import TYPE_CHECKING, Optional, Union, final
+from typing import Optional, Union, final
 
 import torch
 
@@ -21,10 +21,6 @@ if HAS_TRITON:
 
 from .punica_base import PunicaWrapperBase
 
-if TYPE_CHECKING:
-    # avoid circuit import
-    from vllm.lora.models import LongContextLoRAContext
-
 
 @final
 class PunicaWrapperGPU(PunicaWrapperBase):
@@ -55,20 +51,13 @@ class PunicaWrapperGPU(PunicaWrapperBase):
                                                        max_num_prompts,
                                                        device=device)
 
-    def update_metadata(
-            self,
-            mapping: LoRAMapping,
-            lora_index_to_id: list[Optional[int]],
-            max_loras: int,
-            vocab_size: int,
-            extra_vocab_size: int,
-            long_lora_context: Optional["LongContextLoRAContext"] = None,
-            **kwargs):
+    def update_metadata(self, mapping: LoRAMapping,
+                        lora_index_to_id: list[Optional[int]], max_loras: int,
+                        vocab_size: int, extra_vocab_size: int, **kwargs):
 
         self.is_prefill = mapping.is_prefill
         self._update_base_metadata(mapping, lora_index_to_id, max_loras,
-                                   vocab_size, extra_vocab_size,
-                                   long_lora_context)
+                                   vocab_size, extra_vocab_size)
 
         # Prepare cuda kernel metadata tensors
         self.token_mapping_meta.prepare_tensors(self.token_lora_indices)