fix(weave): add attributes to langchain integration calls

[parambharat]

WB-22340

Description

This PR enables to attribute tracking in the langchain tracer integration. This simplest way to highlight the fix is the following code snippet. Now we should be able to track the {"tag": "good_tag"} attribute in trace op for the the langchain calls.

import weave
from langchain_core.prompts import PromptTemplate
from langchain_openai import ChatOpenAI

# Initialize Weave with your project name
weave.init("langchain_test")

llm = ChatOpenAI()
prompt = PromptTemplate.from_template("1 + {number} = ")

llm_chain = prompt | llm
with weave.attributes({"tag": "good_tag"}):
    output = llm_chain.invoke({"number": 2})

print(output)

Testing

Included vcr cassette and unit test to verify that the user-added attributes are present in the call attributes.

[circle-job-mirror]

Preview this PR with FeatureBee: https://beta.wandb.ai/?betaVersion=85b9c1f2bdb429af4a735b828b6935cb8b70049b

[andrewtruong]

please remove fingerprints

[parambharat]

👍 removed

[ayulockin]

@parambharat can you make the CI green and merge this PR.

[parambharat]

@ayulockin, The CI is now 🟢 . But I can't merge the PR since I don't have write access to the repo 🤔