Community: Fuse HuggingFace Endpoint-related classes into one (#17254)

## Description
Fuse HuggingFace Endpoint-related classes into one:
-
[HuggingFaceHub](5ceaf784f3/libs/community/langchain_community/llms/huggingface_hub.py)
-
[HuggingFaceTextGenInference](5ceaf784f3/libs/community/langchain_community/llms/huggingface_text_gen_inference.py)
- and
[HuggingFaceEndpoint](5ceaf784f3/libs/community/langchain_community/llms/huggingface_endpoint.py)

Are fused into
- HuggingFaceEndpoint

## Issue
The deduplication of classes was creating a lack of clarity, and
additional effort to develop classes leads to issues like [this
hack](5ceaf784f3/libs/community/langchain_community/llms/huggingface_endpoint.py (L159)).

## Dependancies

None, this removes dependancies.

## Twitter handle

If you want to post about this: @AymericRoucher

---------

Co-authored-by: Bagatur <baskaryan@gmail.com>

libs/community/tests/integration_tests/llms/test_huggingface_endpoint.py

@@ -1,6 +1,5 @@
-"""Test HuggingFace API wrapper."""
+"""Test HuggingFace Endpoints."""
 
-import unittest
 from pathlib import Path
 
 import pytest
@@ -10,42 +9,9 @@ from langchain_community.llms.loading import load_llm
 from tests.integration_tests.llms.utils import assert_llm_equality
 
 
-@unittest.skip(
-    "This test requires an inference endpoint. Tested with Hugging Face endpoints"
-)
-def test_huggingface_endpoint_text_generation() -> None:
-    """Test valid call to HuggingFace text generation model."""
-    llm = HuggingFaceEndpoint(
-        endpoint_url="", task="text-generation", model_kwargs={"max_new_tokens": 10}
-    )
-    output = llm("Say foo:")
-    print(output)  # noqa: T201
-    assert isinstance(output, str)
-
-
-@unittest.skip(
-    "This test requires an inference endpoint. Tested with Hugging Face endpoints"
-)
-def test_huggingface_endpoint_text2text_generation() -> None:
-    """Test valid call to HuggingFace text2text model."""
-    llm = HuggingFaceEndpoint(endpoint_url="", task="text2text-generation")
-    output = llm("The capital of New York is")
-    assert output == "Albany"
-
-
-@unittest.skip(
-    "This test requires an inference endpoint. Tested with Hugging Face endpoints"
-)
-def test_huggingface_endpoint_summarization() -> None:
-    """Test valid call to HuggingFace summarization model."""
-    llm = HuggingFaceEndpoint(endpoint_url="", task="summarization")
-    output = llm("Say foo:")
-    assert isinstance(output, str)
-
-
 def test_huggingface_endpoint_call_error() -> None:
     """Test valid call to HuggingFace that errors."""
-    llm = HuggingFaceEndpoint(model_kwargs={"max_new_tokens": -1})
+    llm = HuggingFaceEndpoint(endpoint_url="", model_kwargs={"max_new_tokens": -1})
     with pytest.raises(ValueError):
         llm("Say foo:")
 
@@ -58,3 +24,58 @@ def test_saving_loading_endpoint_llm(tmp_path: Path) -> None:
     llm.save(file_path=tmp_path / "hf.yaml")
     loaded_llm = load_llm(tmp_path / "hf.yaml")
     assert_llm_equality(llm, loaded_llm)
+
+
+def test_huggingface_text_generation() -> None:
+    """Test valid call to HuggingFace text generation model."""
+    llm = HuggingFaceEndpoint(repo_id="gpt2", model_kwargs={"max_new_tokens": 10})
+    output = llm("Say foo:")
+    print(output)  # noqa: T201
+    assert isinstance(output, str)
+
+
+def test_huggingface_text2text_generation() -> None:
+    """Test valid call to HuggingFace text2text model."""
+    llm = HuggingFaceEndpoint(repo_id="google/flan-t5-xl")
+    output = llm("The capital of New York is")
+    assert output == "Albany"
+
+
+def test_huggingface_summarization() -> None:
+    """Test valid call to HuggingFace summarization model."""
+    llm = HuggingFaceEndpoint(repo_id="facebook/bart-large-cnn")
+    output = llm("Say foo:")
+    assert isinstance(output, str)
+
+
+def test_huggingface_call_error() -> None:
+    """Test valid call to HuggingFace that errors."""
+    llm = HuggingFaceEndpoint(repo_id="gpt2", model_kwargs={"max_new_tokens": -1})
+    with pytest.raises(ValueError):
+        llm("Say foo:")
+
+
+def test_saving_loading_llm(tmp_path: Path) -> None:
+    """Test saving/loading an HuggingFaceEndpoint LLM."""
+    llm = HuggingFaceEndpoint(repo_id="gpt2", model_kwargs={"max_new_tokens": 10})
+    llm.save(file_path=tmp_path / "hf.yaml")
+    loaded_llm = load_llm(tmp_path / "hf.yaml")
+    assert_llm_equality(llm, loaded_llm)
+
+
+def test_invocation_params_stop_sequences() -> None:
+    llm = HuggingFaceEndpoint()
+    assert llm._default_params["stop_sequences"] == []
+
+    runtime_stop = None
+    assert llm._invocation_params(runtime_stop)["stop_sequences"] == []
+    assert llm._default_params["stop_sequences"] == []
+
+    runtime_stop = ["stop"]
+    assert llm._invocation_params(runtime_stop)["stop_sequences"] == ["stop"]
+    assert llm._default_params["stop_sequences"] == []
+
+    llm = HuggingFaceEndpoint(stop_sequences=["."])
+    runtime_stop = ["stop"]
+    assert llm._invocation_params(runtime_stop)["stop_sequences"] == [".", "stop"]
+    assert llm._default_params["stop_sequences"] == ["."]