sortby(dim) removes dimension from an array if dim is a repeated dimension

[floogit]

Related to #1378. As mentioned there, xarray sometimes seems to have problems with repeated dimensions. sortby() such a dimension seems to be an example:

ds = xr.DataArray(np.eye(2), coords={'alt': [100, 200], 'alt': [100, 200], }, 
    dims=('alt', 'alt'), name='cov').to_dataset()

print(ds)
<xarray.Dataset>
Dimensions:  (alt: 2)
Coordinates:
  * alt      (alt) int64 100 200
Data variables:
    cov      (alt, alt) float64 1.0 0.0 0.0 1.0

print(ds.sortby('alt'))
<xarray.Dataset>
Dimensions:  (alt: 2)
Coordinates:
  * alt      (alt) int64 100 200
Data variables:
    cov      (alt) float64 1.0 1.0

Output of xr.show_versions()

commit: None
python: 2.7.13.final.0
python-bits: 64
OS: Linux
OS-release: 4.9.0-8-amd64
machine: x86_64
processor:
byteorder: little
LC_ALL: None
LANG: en_US.UTF-8
LOCALE: en_US.UTF-8

xarray: 0.10.9
pandas: 0.23.4
numpy: 1.15.4
scipy: 1.1.0
netCDF4: 1.2.8
h5netcdf: None
h5py: None
Nio: None
zarr: None
cftime: 1.0.0
PseudonetCDF: None
rasterio: 0.36.0
iris: None
bottleneck: 1.2.1
cyordereddict: 1.0.0
dask: 0.20.2
distributed: None
matplotlib: 2.2.3
cartopy: 0.16.0
seaborn: 0.8.1
setuptools: 39.2.0
pip: 18.1
conda: None
pytest: 3.8.0
IPython: 5.8.0
sphinx: None

[shoyer]

Thanks for the report. Let's consolidate discussion over in #1378. Help would definitely be appreciated on this -- a great start would be more comprehensive tests (we could merge them marked as xfail).