gblinear is not supported

[Ying456123]

@vruusmann
I'm converting a xgboost model to a PMML model, and find it is still not supported by jpmml-xgboost. Could you help to support it when you have time? Very appreciated.
The model:

from xgboost import XGBClassifier
from sklearn.datasets import load_digits
from sklearn.model_selection import train_test_split
digits = load_digits()
X = digits['data']
y = digits['target']
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)

model = XGBClassifier(verbosity=3, objective='binary:logitraw', booster='gblinear',    reg_alpha=0.0001,
                      reg_lambda=0.98, updater='coord_descent',
                      learning_rate=0.2,  # eta
                      gamma=0.1,
                      n_estimators=20,
                      max_delta_step=3,
                      max_depth=8, 
                      min_child_weight=2, 
                      subsample=0.8, 
                      colsample_bytree=0.5,
                      colsample_bylevel=0.5,
                      colsample_bynode=0.5,
                      scale_pos_weight=1.2, 
                      random_state=27,
                      base_score=0.6,
                      n_jobs=-1)
model.fit(X_train, y_train, eval_metric='error')