Revert "[Bugfix] Fix block size validation (vllm-project#10938)"

This reverts commit 69ba344.

Signed-off-by: Konrad Zawora <kzawora@habana.ai>

vllm/engine/arg_utils.py

@@ -423,7 +423,7 @@ def add_cli_args(parser: FlexibleArgumentParser) -> FlexibleArgumentParser:
         parser.add_argument('--block-size',
                             type=int,
                             default=EngineArgs.block_size,
-                            choices=[8, 16, 32],
+                            choices=[8, 16, 32, 64, 128],
                             help='Token block size for contiguous chunks of '
                             'tokens. This is ignored on neuron devices and '
                             'set to max-model-len')