fix: is_async not passed for llm generation in context recall

[jokokojote]

In _context_recall.py the is_async flag is not passed to the llm.generate function, like it is done for the other metrics which leads to problems when utilising other APIs then OpenAI.

Example:

from ragas.metrics import context_recall
from datasets import Dataset
from ragas import evaluate

import os
os.environ["OPENAI_API_KEY"] = ""
os.environ["RAGAS_DO_NOT_TRACK"] = "true"

questions = [
    "Where is France and what is it’s capital?",
  ]

ground_truths = [
  "France is in Western Europe and its capital is Paris.",
  ]

contexts =[
    [  
        "The country is also renowned for its wines and sophisticated cuisine. Lascaux’s ancient cave drawings, Lyon’s Roman theater and", 
        "The country is also renowned for its wines and sophisticated cuisine. Lascaux’s ancient cave drawings, Lyon’s Roman theater and", 
    ]
]

data = {
    "question": questions,
    "contexts": contexts,
    "ground_truth": ground_truths
}

# Convert dict to dataset
dataset = Dataset.from_dict(data)
dataset


# OpenAI works
result = evaluate(
    dataset = dataset, 
    metrics=[
        context_recall,
    ],
) 

print(result.to_pandas())


# Bedrock does not work
os.environ['AWS_ACCESS_KEY_ID'] = ''
os.environ['AWS_SECRET_ACCESS_KEY'] = ''
from langchain_community.llms.bedrock import Bedrock
llm = Bedrock(model_id="anthropic.claude-v2:1")

result = evaluate(
    dataset = dataset, 
    metrics=[
        context_recall,
    ],
    llm=llm
) # not working, timeout, error: "Streaming must be set to True for async operations."

print(result.to_pandas())

[shahules786]

thanks bro @jokokojote