DMatrix fails on Pandas' nullable floats

[Ark-kun]

XGBoost 1.6.1
Pandas: 1.4.1

Final training data information:
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 10000 entries, 0 to 9999
Data columns (total 9 columns):
 #   Column                  Non-Null Count  Dtype   
---  ------                  --------------  -----   
 0   class                   10000 non-null  Int64   
 1   trip_seconds            9996 non-null   Int64   
 2   trip_miles              10000 non-null  Float64 
 3   pickup_community_area   9418 non-null   Int64   
 4   dropoff_community_area  9267 non-null   Int64   
 5   fare                    10000 non-null  Float64 
 6   tolls                   8920 non-null   Int64   
 7   extras                  10000 non-null  Float64 
 8   company                 10000 non-null  category
dtypes: Float64(3), Int64(5), category(1)
memory usage: 714.4 KB
Traceback (most recent call last):
  File "/tmp/tmp.ap5suk4c6n", line 116, in <module>
    _outputs = xgboost_train(**_parsed_args)
  File "/tmp/tmp.ap5suk4c6n", line 67, in xgboost_train
    training_data = xgboost.DMatrix(
  File "/usr/local/lib/python3.10/site-packages/xgboost/core.py", line 532, in inner_f
    return f(**kwargs)
  File "/usr/local/lib/python3.10/site-packages/xgboost/core.py", line 643, in __init__
    handle, feature_names, feature_types = dispatch_data_backend(
  File "/usr/local/lib/python3.10/site-packages/xgboost/data.py", line 896, in dispatch_data_backend
    return _from_pandas_df(data, enable_categorical, missing, threads,
  File "/usr/local/lib/python3.10/site-packages/xgboost/data.py", line 345, in _from_pandas_df
    data, feature_names, feature_types = _transform_pandas_df(
  File "/usr/local/lib/python3.10/site-packages/xgboost/data.py", line 283, in _transform_pandas_df
    _invalid_dataframe_dtype(data)
  File "/usr/local/lib/python3.10/site-packages/xgboost/data.py", line 247, in _invalid_dataframe_dtype
    raise ValueError(msg)
ValueError: DataFrame.dtypes for data must be int, float, bool or category.  When
categorical type is supplied, DMatrix parameter `enable_categorical` must
be set to `True`. Invalid columns:trip_miles, fare, extras, company

#7760 only added support for nullable integers and booleans, but not floats.

P.S. Also the _invalid_dataframe_dtype incorrectly includes categorical columns despite enable_categorical set to True.

[trivialfis]

Thank you for raising the issue! Would you like to open a PR for the fix?