[Docker] bump up neuron sdk v2.21 (#11593)

Signed-off-by: Liangfu Chen <liangfc@amazon.com>

vllm/_custom_ops.py

@@ -23,8 +23,7 @@ with contextlib.suppress(ImportError):
     import vllm._moe_C  # noqa: F401
     supports_moe_ops = True
 
-# neuron has torch version that doesn't even have impl_abstract
-if TYPE_CHECKING or current_platform.is_neuron():
+if TYPE_CHECKING:
 
     def register_fake(fn):
         return lambda name: fn

vllm/triton_utils/importing.py

@@ -8,7 +8,6 @@ logger = init_logger(__name__)
 HAS_TRITON = (
     find_spec("triton") is not None
     and not current_platform.is_xpu()  # Not compatible
-    and not current_platform.is_neuron()  # neuron has too old torch
 )
 
 if not HAS_TRITON: