Fix TypeError in aa_omniscience_prompt

[pjavanrood]

Describe the bug

The aa_omniscience prompt function was missing the task_name argument in its signature or call, causing a TypeError when the evaluator attempted to pass the task name.

To Reproduce

task = "aa_omniscience|5"

pipeline = Pipeline(
    tasks=task,
    pipeline_parameters=pipeline_params,
    evaluation_tracker=evaluation_tracker,
    model_config=model_config,
)

pipeline.evaluate()
pipeline.save_and_push_results()
pipeline.show_results()

    140 # We init tasks first to fail fast if one is badly defined
    141 self._init_random_seeds()
--> 142 self._init_tasks_and_requests(tasks=tasks)
    144 self.model_config = model_config
    145 self.accelerator, self.parallel_context = self._init_parallelism_manager()
...
--> 309 doc = self.formatter(item, self.name)
    310 # Skip if formatter returns None (e.g., to filter out certain samples)
    311 if doc is None or doc == []:

TypeError: aa_omniscience_prompt() takes 1 positional argument but 2 were given

Expected behavior

The function should accept task_name as an optional argument and return a Doc object containing that task_name.

Version info

• OS: mac
• Lighteval version: main (local development)