Support for additional LLM generators in Langfuse integration

[mrm1001]

At the moment, we only support OpenAI generator, but there has been demand in the community to use other models, such as Anthropic's Claude LLMs.

Below is an example from Ardaric in discord:

import os
from haystack import Pipeline
from haystack.components.builders import PromptBuilder
from haystack_integrations.components.generators.anthropic import AnthropicGenerator

from dotenv import load_dotenv

def make_pipeline():
    from haystack_integrations.components.connectors.langfuse import LangfuseConnector
    pipe = Pipeline()

    extract_prompt_builder = PromptBuilder(template=extract_question_prompt)
    extractor_llm = AnthropicGenerator(model="claude-3-haiku-20240307")

    pipe.add_component("tracer", LangfuseConnector("Test Logging"))
    pipe.add_component(instance=extract_prompt_builder, name="extract_prompt_builder")
    pipe.add_component(instance=extractor_llm, name="extractor_llm")

    pipe.connect("extract_prompt_builder", "extractor_llm")

    return pipe

if __name__ == "__main__":
    load_dotenv()
    print(os.getenv('LANGFUSE_SECRET_KEY'), os.getenv('LANGFUSE_PUBLIC_KEY'), os.getenv('LANGFUSE_HOST'), os.getenv('HAYSTACK_CONTENT_TRACING_ENABLED'))
    # os.environ['LANGFUSE_DEBUG'] = 'True'
    extract_question_prompt = "{{batch}} | How old is John?"
    pipe = make_pipeline()
    result = pipe.run(
        {"extract_prompt_builder": {"batch": "John is 34 years old"}}
    )
    print(result)