dataset list always returns [] (load_dataset returns a dict, not a model)

[roll]

dataset list always returns an empty list because load_dataset returns a dict.

Environment

• fairspec 0.1.4 (installed from PyPI)
• Python 3.12.3, Linux
• Reproduced in a clean virtualenv against the published wheel.

Repro

cat > dataset.json <<'JSON'
{ "resources": [ { "name": "products", "data": "products.csv" } ] }
JSON
fairspec dataset validate dataset.json --json   # {"valid": true, "errors": []}
fairspec dataset list dataset.json --json        # []   <-- expected ["products"]

Root cause
commands/dataset/list_.py does:

dataset = load_dataset(path)
resource_names = [
    resource.name or infer_resource_name(resource)
    for resource in getattr(dataset, "resources", None) or []
]

but load_dataset() returns a plain dict, so getattr(dict_obj, "resources", None) is always None → the comprehension iterates over [].

from fairspec import load_dataset
ds = load_dataset("dataset.json")
print(type(ds).__name__)            # dict
print(getattr(ds, "resources", None))  # None

The dataset script docs (docs/terminal/dataset.md) also show dataset.resources[0].name, which would fail for the same dict-vs-model reason.

Expected: dataset list returns the resource names (e.g. ["products"]).