`ai_track` uses multiple values for keyword argument `'op'`

[mshavliuk]

How do you use Sentry?

Sentry Saas (sentry.io)

Version

2.33.0

Steps to Reproduce

import os

import sentry_sdk
from langchain_aws import ChatBedrockConverse
from sentry_sdk.ai.monitoring import ai_track
from sentry_sdk.integrations.langchain import LangchainIntegration

sentry_sdk.init(
    dsn=os.environ["SENTRY_DSN"],
    environment="localhost",
    traces_sample_rate=1.0,
    integrations=[LangchainIntegration()],
)

@ai_track("Bug demo", op="ai.run")
def run_llm():
    llm = ChatBedrockConverse(
        region_name="eu-west-1",
        aws_access_key_id=os.environ["AWS_ACCESS_KEY_ID"],
        aws_secret_access_key=os.environ["AWS_SECRET_ACCESS_KEY"],
        model="eu.amazon.nova-lite-v1:0",
        max_tokens=1,
    )
    llm.invoke('hello')


if __name__ == "__main__":
    run_llm()

Expected Result

The llm call is traced with "ai.run" operation name

Actual Result

Traceback (most recent call last):
  File ".../sentry_bug.py", line 29, in <module>
    run_llm()
  File ".../.venv/lib/python3.12/site-packages/sentry_sdk/ai/monitoring.py", line 37, in sync_wrapped
    with start_span(name=description, op=op, **span_kwargs) as span:
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: sentry_sdk.api.start_span() got multiple values for keyword argument 'op'

The issue is caused in the following code:

sentry-python/sentry_sdk/ai/monitoring.py

Lines 35 to 37 in 34dcba4

	op = span_kwargs.get("op", "ai.run" if curr_pipeline else "ai.pipeline")
	with start_span(name=description, op=op, **span_kwargs) as span:

[linear]

PY-1778 `ai_track` uses multiple values for keyword argument `'op'`