Pandas groupby(..).apply() has inconsistent behavior when there is only one group

Code Sample, a copy-pastable example if possible

df = pd.DataFrame({'a':['foo', 'foo', 'bar'], 'b':[1,1,1]}).set_index('a')
def f(df):
    print '============'
    print df
    print
    print(pd.DataFrame(df.groupby(level='a').apply(lambda df: df['b'].cumsum())))

f(df)
f(df[df.index == 'foo'])

Actual Output

============
     b
a     
foo  1
foo  1
bar  1

         b
a   a     
bar bar  1
foo foo  1
    foo  2
============
     b
a     
foo  1
foo  1

a    foo  foo
a            
foo    1    2

Expected Output

============
     b
a     
foo  1
foo  1
bar  1

         b
a   a     
bar bar  1
foo foo  1
    foo  2
============
     b
a     
foo  1
foo  1

         b
a   a     
foo foo  1
    foo  2

output of pd.show_versions()

INSTALLED VERSIONS

commit: None python: 2.7.10.final.0 python-bits: 64 OS: Darwin OS-release: 15.5.0 machine: x86_64 processor: i386 byteorder: little LC_ALL: None LANG: en_US.UTF-8

pandas: 0.17.1 nose: 1.3.4 pip: 8.1.1 setuptools: 18.0.1 Cython: None numpy: 1.9.2 scipy: 0.15.1 statsmodels: 0.6.1 IPython: 4.0.0 sphinx: None patsy: 0.4.1 dateutil: 2.2 pytz: 2014.10 blosc: None bottleneck: None tables: None numexpr: None matplotlib: 1.4.3 openpyxl: None xlrd: None xlwt: None xlsxwriter: None lxml: None bs4: None html5lib: None httplib2: None apiclient: None sqlalchemy: 1.0.12 pymysql: None psycopg2: None Jinja2: None

Comment From: jreback

tracked in #13056