[Feature] support logits processors (#4515)

* [feat] provide an interface for logits processors and a builtin LogitBiasLogitsProcessor

* [chore] fix code style

* [fix] add unit test & fix existing bugs

* [feat] add engine/worker arg --logits-processors

* [fix] redefine user args as logits_processors_args and fix some bugs

* [fix] fix test_sampler

* Update fastdeploy/model_executor/logits_processor/builtin.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* Update fastdeploy/model_executor/logits_processor/__init__.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* Update tests/model_executor/test_logits_processor.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* [fix] fix typo

* Update fastdeploy/engine/sampling_params.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* [fix] fix bracelet

* [chore] redefine logits processor interface: pass the entire share_inputs into LP, do not copy share_inputs and logits

* [doc] add docs

* [fix] fix logit bias processor not applied when decoding is too fast & add docs and tests

* [fix] fix redundant code

* [feat] skip apply() if no bias is specified

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

fastdeploy/engine/args_utils.py

@@ -422,6 +422,16 @@ class EngineArgs:
     Flag to specify the dtype of lm_head as FP32. Default is False (Using model default dtype).
     """
 
+    logits_processors: Optional[List[str]] = None
+    """
+    A list of FQCNs (Fully Qualified Class Names) of logits processors supported by the service.
+    A fully qualified class name (FQCN) is a string that uniquely identifies a class within a Python module.
+
+    - To enable builtin logits processors, add builtin module paths and class names to the list. Currently support:
+        - fastdeploy.model_executor.logits_processor:LogitBiasLogitsProcessor
+    - To enable custom logits processors, add your dotted paths to module and class names to the list.
+    """
+
     def __post_init__(self):
         """
         Post-initialization processing to set default tokenizer if not provided.
@@ -687,6 +697,13 @@ class EngineArgs:
             default=EngineArgs.lm_head_fp32,
             help="Specify the dtype of lm_head weight as float32.",
         )
+        model_group.add_argument(
+            "--logits-processors",
+            type=str,
+            nargs="+",
+            default=EngineArgs.logits_processors,
+            help="FQCNs (Fully Qualified Class Names) of logits processors supported by the service.",
+        )
 
         # Parallel processing parameters group
         parallel_group = parser.add_argument_group("Parallel Configuration")