Ensure lists cannot be passed into DMatrix

The documentation does not include lists as an allowed type for the data inputted into DMatrix. Despite this, a list can be passed in without an error. This change would prevent a list form being passed in directly.

I experienced an issue where passing in a list vs a np.array resulted in different predictions (sometimes over 10% relative difference) for the same data. Though these differences were infrequent (~1.5% of cases tested), in certain applications this could cause serious issues.

python-package/xgboost/core.py

@@ -381,7 +381,9 @@ def __init__(self, data, label=None, missing=None,
         label = _maybe_dt_array(label)
         weight = _maybe_dt_array(weight)
 
-        if isinstance(data, STRING_TYPES):
+        if isinstance(data, list):
+            raise TypeError('can not initialize DMatrix from list)
+        elif isinstance(data, STRING_TYPES):
             self.handle = ctypes.c_void_p()
             _check_call(_LIB.XGDMatrixCreateFromFile(c_str(data),
                                                      ctypes.c_int(silent),