fillna can't process np.float16 type

[MarsGarai]

Code Sample, a copy-pastable example if possible

p_s = pd.Series(data=[47.97,   np.NAN,  np.NAN, 47.16, 47.97], dtype=np.float16, 
                index=pd.DatetimeIndex(['2015-07-24', '2015-07-25', '2015-07-26', \
                                                          '2015-07-27','2015-07-28']))
p_s
Out:
2015-07-24    47.96875
2015-07-25         NaN
2015-07-26         NaN
2015-07-27    47.15625
2015-07-28    47.96875
dtype: float16

Problem description

p_s.fillna(method='ffill')
get the error:
ValueError: Invalid dtype for pad_2d [float16]
But:
p_s.astype(np.float64).fillna(method='ffill')
or:
p_s.astype(np.float32).fillna(method='ffill')
works well:
2015-07-24 47.96875
2015-07-25 47.96875
2015-07-26 47.96875
2015-07-27 47.15625
2015-07-28 47.96875

Expected Output

p_s..fillna(method='ffill')
Out:
2015-07-24 47.96875
2015-07-25 47.96875
2015-07-26 47.96875
2015-07-27 47.15625
2015-07-28 47.96875
dtype: float16

Output of pd.show_versions()

[paste the output of pd.show_versions() here below this line]

1. On WINDOWS
   INSTALLED VERSIONS

---

commit: None
python: 3.6.3.final.0
python-bits: 64
OS: Windows
OS-release: 10
machine: AMD64
processor: Intel64 Family 6 Model 58 Stepping 9, GenuineIntel
byteorder: little
LC_ALL: None
LANG: None
LOCALE: None.None

pandas: 0.20.3
pytest: 3.2.1
pip: 9.0.1
setuptools: 36.5.0.post20170921
Cython: 0.26.1
numpy: 1.14.0
scipy: 1.0.0
xarray: None
IPython: 6.1.0
sphinx: 1.6.3
patsy: 0.4.1
dateutil: 2.6.1
pytz: 2017.2
blosc: None
bottleneck: 1.2.1
tables: 3.4.2
numexpr: 2.6.2
feather: None
matplotlib: 2.1.0
openpyxl: 2.4.8
xlrd: 1.1.0
xlwt: 1.3.0
xlsxwriter: 1.0.2
lxml: 4.1.0
bs4: 4.6.0
html5lib: 0.9999999
sqlalchemy: 1.1.13
pymysql: None
psycopg2: None
jinja2: 2.9.6
s3fs: None
pandas_gbq: None
pandas_datareader: None

2. On UBUNTU 16.04
   INSTALLED VERSIONS

---

commit: None
python: 3.6.2.final.0
python-bits: 64
OS: Linux
OS-release: 4.4.0-108-generic
machine: x86_64
processor: x86_64
byteorder: little
LC_ALL: None
LANG: ru_RU.UTF-8
LOCALE: ru_RU.UTF-8

pandas: 0.22.0
pytest: 3.2.1
pip: 9.0.1
setuptools: 36.5.0.post20170921
Cython: 0.26.1
numpy: 1.14.0
scipy: 1.0.0
pyarrow: None
xarray: None
IPython: 6.1.0
sphinx: 1.6.3
patsy: 0.4.1
dateutil: 2.6.1
pytz: 2017.2
blosc: None
bottleneck: 1.2.1
tables: 3.4.2
numexpr: 2.6.2
feather: None
matplotlib: 2.0.2
openpyxl: 2.4.8
xlrd: 1.1.0
xlwt: 1.3.0
xlsxwriter: 0.9.8
lxml: 3.8.0
bs4: 4.6.0
html5lib: 0.999999999
sqlalchemy: 1.1.13
pymysql: None
psycopg2: None
jinja2: 2.9.6
s3fs: None
fastparquet: None
pandas_gbq: None
pandas_datareader: None

[jreback]

float16 have almost no support anywhere as a first class type.

[jreback]

we have a master issue about this

[gustavz]

So this issue won't be addressed in future updates?

[jreback]

see #9220

it would take pull requests from the community
pandas is all volunteer and there are 3000 issues open