Fix AttributeError in onnxruntime train_unconditional (args.report_to → args.logger)

[Ricardo-M-L]

What does this PR do?

The ORT variant of the unconditional training script,
examples/research_projects/onnxruntime/unconditional_image_generation/train_unconditional.py,
defines the tracker flag as --logger on the parser:

parser.add_argument(
    \"--logger\",
    type=str,
    default=\"tensorboard\",
    choices=[\"tensorboard\", \"wandb\"],
    ...
)

…but main() references a non-existent args.report_to twice:

if args.report_to == \"wandb\" and args.hub_token is not None:
    raise ValueError(
        \"You cannot use both --report_to=wandb and --hub_token ...\"
        \" Please use \`hf auth login\` to authenticate with the Hub.\"
    )

...

accelerator = Accelerator(
    gradient_accumulation_steps=args.gradient_accumulation_steps,
    mixed_precision=args.mixed_precision,
    log_with=args.report_to,
    project_config=accelerator_project_config,
)

Because no --report_to flag is registered, the args namespace never carries that attribute, so both references raise AttributeError: 'Namespace' object has no attribute 'report_to' at the very top of main(). The script can't run end-to-end.

The sibling non-ORT script examples/unconditional_image_generation/train_unconditional.py uses args.logger throughout. This PR aligns the ORT variant with that:

• args.report_to → args.logger at both call sites
• --report_to=wandb → --logger=wandb in the error message text

Trivial one-word fix; no behavior change beyond letting the script actually start.

Before submitting

• This PR fixes a typo or improves the docs.
• Did you read the contributor guideline?
• Did you make sure to update the documentation with your changes? (N/A — error-message text only)
• Did you write any new necessary tests? (N/A — trivially verifiable attribute name)

Who can review?

@sayakpaul @stevhliu