[Serve.llm] Remove ImageRetriever class and related tests from the LLM deployment module. (#53980)

Signed-off-by: Kourosh Hakhamaneshi <kourosh@anyscale.com>

Author: kouroshHakha

ci/lint/pydoclint-baseline.txt

@@ -1612,8 +1612,6 @@ python/ray/llm/_internal/serve/configs/prompt_formats.py
     DOC201: Method `Image.check_image_url` does not have a return section in docstring
 --------------------
 python/ray/llm/_internal/serve/deployments/llm/llm_server.py
-    DOC101: Method `LLMServer.__init__`: Docstring contains fewer arguments than in function signature.
-    DOC103: Method `LLMServer.__init__`: Docstring arguments are different from function arguments. (Or could be other formatting issues: https://jsh9.github.io/pydoclint/violation_codes.html#notes-on-doc103 ). Arguments in the function signature but not in the docstring: [engine_cls: Optional[Type[LLMEngine]], image_retriever_cls: Optional[Type[ImageRetriever]], model_downloader: Optional[LoraModelLoader]].
     DOC402: Method `LLMServer.embeddings` has "yield" statements, but the docstring does not have a "Yields" section
     DOC404: Method `LLMServer.embeddings` yield type(s) in docstring not consistent with the return annotation. Return annotation exists, but docstring "yields" section does not exist or has 0 type(s).
 --------------------

python/ray/llm/_internal/serve/deployments/llm/image_retriever.py

@@ -45,7 +45,6 @@
     LLMConfig,
     LLMRawResponse,
 )
-from ray.llm._internal.serve.deployments.llm.image_retriever import ImageRetriever
 from ray.llm._internal.serve.deployments.llm.llm_engine import LLMEngine
 from ray.llm._internal.serve.deployments.llm.multiplex.lora_model_loader import (
     LoraModelLoader,
@@ -409,14 +408,12 @@ async def process_completions(
 
 class LLMServer(_LLMServerBase):
     _default_engine_cls = VLLMEngine
-    _default_image_retriever_cls = ImageRetriever
 
     async def __init__(
         self,
         llm_config: LLMConfig,
         *,
         engine_cls: Optional[Type[LLMEngine]] = None,
-        image_retriever_cls: Optional[Type[ImageRetriever]] = None,
         model_downloader: Optional[LoraModelLoader] = None,
     ):
         """Constructor of LLMServer.
@@ -426,14 +423,10 @@ async def __init__(
 
         Args:
             llm_config: LLMConfig for the model.
-
-        Keyword Args:
-            engine_cls: Dependency injection for the vllm engine class. Defaults to
-                `VLLMEngine`.
-            image_retriever_cls: Dependency injection for the image retriever class.
-                Defaults to `ImageRetriever`.
-            model_downloader: Dependency injection for the model downloader object.
-                Defaults to be initialized with `LoraModelLoader`.
+            engine_cls: Dependency injection for the vllm engine class.
+                Defaults to `VLLMEngine`.
+            model_downloader: Dependency injection for the model downloader
+                object. Defaults to be initialized with `LoraModelLoader`.
         """
         await super().__init__(llm_config)
 
@@ -443,14 +436,6 @@ async def __init__(
             self.engine = self._engine_cls(self._llm_config)
             await asyncio.wait_for(self._start_engine(), timeout=ENGINE_START_TIMEOUT_S)
 
-        # TODO (Kourosh): I think we can completely remove image retriever.
-        # It was missed to get removed.
-        self.image_retriever = (
-            image_retriever_cls()
-            if image_retriever_cls
-            else self._default_image_retriever_cls()
-        )
-
         multiplex_config = self._llm_config.multiplex_config()
         if model_downloader:
             self.model_downloader = model_downloader

python/ray/llm/_internal/serve/deployments/llm/llm_server.py

@@ -15,7 +15,6 @@
 from ray.llm._internal.serve.deployments.routers.router import (
     LLMRouter,
 )
-from ray.llm.tests.serve.mocks.fake_image_retriever import FakeImageRetriever
 from ray.llm.tests.serve.mocks.mock_vllm_engine import MockEchoVLLMEngine
 from ray.serve.handle import DeploymentHandle
 
@@ -68,7 +67,6 @@ def get_mocked_llm_deployments(llm_configs) -> List[DeploymentHandle]:
             deployment.bind(
                 llm_config=llm_config,
                 engine_cls=MockEchoVLLMEngine,
-                image_retriever_cls=FakeImageRetriever,
             )
         )
     return llm_deployments