MultiIndex.remove_unused_levels() fills nans

[toobaz]

Code Sample, a copy-pastable example if possible

In [2]: mi = pd.MultiIndex.from_tuples([('a', 'w'), ('b', 'x'), (np.nan, 'y'), ('d', 'z')])

In [3]: list(mi) # OK
Out[3]: [('a', 'w'), ('b', 'x'), (nan, 'y'), ('d', 'z')]

In [4]: list(mi.remove_unused_levels()) # not OK
Out[4]: [('a', 'w'), ('b', 'x'), ('d', 'y'), ('d', 'z')]

Problem description

The missing value is replaced with another value.

Expected Output

In [3] and In [4] should return the same.

Output of pd.show_versions()

INSTALLED VERSIONS

commit: None
python: 3.5.3.final.0
python-bits: 64
OS: Linux
OS-release: 4.9.0-4-amd64
machine: x86_64
processor:
byteorder: little
LC_ALL: None
LANG: en_GB.UTF-8
LOCALE: en_GB.UTF-8

pandas: 0.22.0.dev0+163.g3b05a60e6.dirty
pytest: 3.0.6
pip: 9.0.1
setuptools: 33.1.1
Cython: 0.25.2
numpy: 1.12.1
scipy: 0.18.1
pyarrow: None
xarray: None
IPython: 5.2.2
sphinx: None
patsy: 0.4.1+dev
dateutil: 2.6.0
pytz: 2016.10
blosc: None
bottleneck: 1.2.0
tables: 3.3.0
numexpr: 2.6.1
feather: 0.3.1
matplotlib: 2.0.0
openpyxl: 2.3.0
xlrd: 1.0.0
xlwt: 1.2.0
xlsxwriter: None
lxml: 3.7.1
bs4: 4.5.3
html5lib: 0.999999999
sqlalchemy: 1.0.15
pymysql: None
psycopg2: None
jinja2: 2.8
s3fs: None
fastparquet: None
pandas_gbq: None
pandas_datareader: None

[jreback]

hmm that seems buggy