missing access weight= in DMatrix for sample weight

[Moelf]

in Python, this is called sample_weight https://xgboost.readthedocs.io/en/stable/python/python_api.html#xgboost.XGBRegressor.fit

looks like XGBoost.jl exposes it via: https://github.com/dmlc/XGBoost.jl/blob/08c38da58050ca2df29f0e5fdcc06ce21de1dcd8/src/dmatrix.jl#L56

Do we have access to this concept in this package?

[Moelf]

I think the right way to do this is to add

+ 229 MMI.supports_weights(::Type{<:XGBoostRegressor}) = true
+ 230 MMI.supports_weights(::Type{<:XGBoostClassifier}) = true
  231

and then also add a 4th argument to:

MLJXGBoostInterface.jl/src/MLJXGBoostInterface.jl

Lines 163 to 166 in 402861a

	function MMI.fit(model::XGBoostClassifier,
	verbosity, # must be here even if unsupported in pkg
	X, y,
	)

?