Does ForestDML support multiple binary treatments?

From the this webpage it reads:

"Our implementation of a Causal Forest is restricted to binary treatment or single-dimensional continuous treatment ..."

If I understand correctly, then it would work if I have multiple binary treatment? However, I tried many ways of specifying and fitting the models and had no luck.

For example, if I use
est = ForestDML(model_t = MultiOutputClassifier(RidgeClassifier()), model_y = Ridge()); est.fit(Y=Y, T=T, X=X, W=W, inference='auto')
(my X array has 2 columns and 60000 rows.; the values of this array are all boolean)
Then I got error:
This method can only be used with single-dimensional continuous treatment or binary categorical treatment.

If I use:
est = ForestDML(model_t = MultiOutputClassifier(RidgeClassifier()), model_y = Ridge(), discrete_treatment=True); est.fit(Y=Y, T=T, X=X, W=W, inference='auto')

Then I got error:
Found input variables with inconsistent numbers of samples: [60000, 120000]

So does ForestDML work for multiple binary treatments? If so, any short example or guide would be really appreciated.

Start with ForestDML's treatment validation and the linked forest documentation, then reproduce both supplied fit calls with the two-column boolean treatment input. Compare the documented binary or single-dimensional treatment constraint with the observed errors; done means establishing whether multiple binary treatments are supported and documenting or scoping the required change.