[Frontend] Add backend-specific options for guided decoding (#13505)

Signed-off-by: Joe Runde <Joseph.Runde@ibm.com>

vllm/config.py

@@ -25,6 +25,7 @@ from vllm.model_executor.layers.quantization import (QUANTIZATION_METHODS,
                                                      get_quantization_config)
 from vllm.model_executor.models import ModelRegistry
 from vllm.platforms import CpuArchEnum
+from vllm.sampling_params import GuidedDecodingParams
 from vllm.tracing import is_otel_available, otel_import_error_traceback
 from vllm.transformers_utils.config import (
     ConfigFormat, get_config, get_hf_image_processor_config,
@@ -2631,7 +2632,9 @@ class DecodingConfig:
 
     def __post_init__(self):
         valid_guided_backends = ['outlines', 'lm-format-enforcer', 'xgrammar']
-        backend = self.guided_decoding_backend
+
+        backend = GuidedDecodingParams(
+            backend=self.guided_decoding_backend).backend_name
         if backend not in valid_guided_backends:
             raise ValueError(f"Invalid guided_decoding_backend '{backend},"
                              f"must be one of {valid_guided_backends}")