Apparently not possible to create models that are both multi class and multi label

I've been looking for ages for a way to train multi label - multi class setfit models and I've reached the conclusion that this is either not supported or there is a desperate need for it to be properly documented. Can I ask if this is a feature that could be considered? I include a minimal example bellow of my failed attempts.

```
from datasets import Dataset
from setfit import SetFitModel, Trainer, TrainingArguments
import pandas as pd

dfA = [["test",["A","x"]],["not",["B","y"]],["yes",["C","z"]]]
dfB = [["test",["A",True]], ["not",["B",True]],["yes",["C",False]]]

args = TrainingArguments(num_epochs=1)

for t in ["multi-output","one-vs-rest","classifier-chain"]:
model = SetFitModel.from_pretrained(
"sentence-transformers/paraphrase-mpnet-base-v2",
multi_target_strategy=t)
for df in [dfA,dfB]:
try:
dataset = Dataset.from_pandas(pd.DataFrame(df,
columns=["text","label"]))
trainer = Trainer(
model=model,
args=args,
train_dataset=dataset
)
trainer.train()
except:
print("fail")
```