[Cleanup] Use tokens_input() for TTS prompt construction (#3227)

Signed-off-by: Nick Cao <ncao@redhat.com>
Co-authored-by: Claude <noreply@anthropic.com>

tests/entrypoints/openai_api/test_serving_speech_voxcpm.py

@@ -109,6 +109,7 @@ class TestVoxCPMServing:
         assert call.kwargs["prompt"] == {
             "prompt_token_ids": [1],
             "additional_information": tts_params,
+            "type": "token",
         }
         assert call.kwargs["output_modalities"] == ["audio"]
 
@@ -140,4 +141,5 @@ class TestVoxCPMServing:
         assert call.kwargs["prompt"] == {
             "prompt_token_ids": [1],
             "additional_information": tts_params,
+            "type": "token",
         }