feat: add progress_format support for machine-readable JSON output

[podarok]

Summary

Adds progress_format support to huggingface_hub, enabling machine-readable JSON progress output similar to huggingface/tokenizers#1921 and huggingface/datasets#7920.

Motivation

When using huggingface_hub in automated pipelines, web backends, or UI applications, it is useful to emit machine-readable progress instead of ANSI progress bars. This PR adds the same progress_format option that was implemented in tokenizers and datasets.

Changes

New Functions

• set_progress_format(format: str): Set global progress format
• get_progress_format() -> str: Get current progress format

Supported Formats

1. "tqdm" (default): Interactive progress bars (current behavior, unchanged)
2. "json": Machine-readable JSON lines to stderr
3. "silent": No output

JSON Format

When progress_format="json", emits JSON every 5% progress change or at completion:

{"stage": "Downloading model.safetensors", "current": 1024, "total": 4096, "percent": 25.0}

Usage Example

from huggingface_hub import hf_hub_download
from huggingface_hub.utils import set_progress_format

# Enable JSON output
set_progress_format("json")

# Progress will now be emitted as JSON lines to stderr
hf_hub_download("gpt2", "config.json")
# Outputs: {"stage":"config.json","current":665,"total":665,"percent":100.0}

Implementation Details

• Suppresses visual output using io.StringIO() when format is "json"
• Keeps progress tracking active (unlike disable=True)
• Emits JSON to stderr every 5% progress change
• Exports new functions from huggingface_hub.utils

Backward Compatibility

• Default behavior is "tqdm" - identical to current behavior
• Existing code works without any changes
• New option is opt-in only

Cross-Reference

This implementation mirrors the approach from:

• huggingface/tokenizers#1921
• huggingface/datasets#7920

[Wauplin]

Hi @podarok, do you have an exact use case for that? If yes, have you checked the tqdm_class argument in hf_hub_download and snapshot_download that allows you to pass a custom tqdm class. This is usually what people use when they want to integrate in an exactly UI/App (see #1110)