fix: pin langchain_core to <3 (explodinggradients#1329)

The new langchain v0.3 will break the current usage of metrics. 

the plan for actions is as follows
1. for ragas<0.2 we will pin langchain_core to <0.3
2. for ragas>0.2 we will directly depend on pydantic>=2

fixes: explodinggradients#1328

fix: column rename (explodinggradients#1324)

fix: make score nested if loop_is_running (explodinggradients#1276)

feat: set and get prompts for metrics (explodinggradients#1259)

```python
from ragas.experimental.metrics._faithfulness import FaithfulnessExperimental, LongFormAnswerPrompt

faithfulness = FaithfulnessExperimental() 
faithfulness.get_prompts()

#{'long_form_answer_prompt': <ragas.experimental.metrics._faithfulness.LongFormAnswerPrompt at 0x7fd7baa8efb0>,
#'nli_statement_prompt': <ragas.experimental.metrics._faithfulness.NLIStatementPrompt at 0x7fd7baa8f010>}

long_form_prompt = LongFormAnswerPrompt()
long_form_prompt.instruction = "my new instruction"

prompts = {"long_form_answer_prompt":long_form_prompt}
faithfulness.set_prompts(**prompts)
```

---------

Co-authored-by: Jithin James <[email protected]>

feat (experimental) added new prompt and metric into `ragas.experimen…

…tal` (explodinggradients#1240)

you can use it like
```py
from ragas.experimental.metrics import FaithfulnessExperimental
from ragas.metrics import faithfulness
from ragas import evaluate


f = FaithfulnessExperimental(llm=LangchainLLMWrapper(gpt4o))
faithfulness.llm = LangchainLLMWrapper(gpt4o)

# row = amnesty_qa["eval"][0]
# await f.ascore(row)
# await faithfulness.ascore(row)

r = evaluate(
    amnesty_qa["eval"].select(range(10)),
    metrics=[f, faithfulness],
    raise_exceptions=True,
    callbacks=[]
)
```

[FIX] - Fix for summarization edge case (explodinggradients#1201)

This PR adds a fix for the issue mentioned in explodinggradients#1108 

However I have a points to discuss @shahules786 :
- I had added `conciseness_score` to penalize long summaries, but I also
do not want to promote very very short and skimpy summaries, need to
find a middle ground.
- Is `averaging` a good way to combine `QA_score` and
`conciseness_score`?
- Ranking based metrics to measure quality of summarization (as
mentioned by shahul in the above issue)

Given the conclusions we reach based on these discussion points, I will
push more commits, let's keep this PR open till we resolve these points.

---------

Co-authored-by: Shahules786 <[email protected]>

fix: added a test for prompts (explodinggradients#1197)

Allow Metric.score to work within an existing asyncio loop (exploding…

…gradients#1161)

I got errors when running metrics within an existing asyncio loop, the
`asyncio.run` part makes the code fail in those cases. Submitting a
bugfix.

---------

Co-authored-by: jjmachan <[email protected]>

fix: temperature parameter in generate_text not ignored. (explodinggr…

…adients#887)

Addresses explodinggradients#886. 

Temperature parameter is not anymore overwritten by calling to
`self.get_temperature(n=n)`. Now it will only call that method if no
parameter was given.

---------

Co-authored-by: Ivan Herreros <[email protected]>

fix: remove `is_async` argument since everything is async now (explod…

…inggradients#1116)