[Intel GPU] Enable DeepSeek V3.2 inference on XPU (#24356)

Signed-off-by: P V R K Jyothendra Varma <polisetty.v.r.k.jyothendra.varma@intel.com>

python/sglang/srt/layers/attention/nsa/nsa_indexer.py

@@ -1062,6 +1062,19 @@ class Indexer(MultiPlatformOp):
             index_k_scale=k_scale,
         )
 
+    def forward_xpu(
+        self,
+        x: torch.Tensor,
+        q_lora: torch.Tensor,
+        positions: torch.Tensor,
+        forward_batch: ForwardBatch,
+        layer_id: int,
+        return_indices: bool = True,
+    ) -> Optional[torch.Tensor]:
+        return self.forward_cuda(
+            x, q_lora, positions, forward_batch, layer_id, return_indices
+        )
+
     def forward_cuda(
         self,
         x: torch.Tensor,

python/sglang/srt/server_args.py

@@ -1707,7 +1707,7 @@ class ServerArgs:
                     self.attention_backend = "nsa"
                     logger.info("Use nsa attention backend for DeepSeek with DSA.")
 
-                if not is_npu():  # CUDA or ROCm GPU
+                if not is_npu() and not is_xpu():  # CUDA or ROCm GPU
                     if self.enable_nsa_prefill_context_parallel:
                         logger.warning(
                             "Context parallel feature is still under experiment. It has only been verified on Hopper platform."