BUG: inconsistent order of evaluation for table error checks

[wasade]

An example of the bug is below. If print statements are added into ErrorProfile.test, it can be shown that there is an unstable order of evaluation. The reason the second example below works is that empty is by default set to ignore and not to raise. However, empty does not presently assert a lack of axis IDs or metadata which is surprising.

$ python
Python 3.5.5 | packaged by conda-forge | (default, Apr  6 2018, 13:43:56)
[GCC 4.2.1 Compatible Apple LLVM 6.1.0 (clang-602.0.53)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import biom
>>> t = biom.Table([], [], ['foo', 'bar'])
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Users/dtmcdonald/ResearchWork/software/biom-format/biom/table.py", line 508, in __init__
    errcheck(self)
  File "/Users/dtmcdonald/ResearchWork/software/biom-format/biom/err.py", line 472, in errcheck
    raise ret
biom.exception.TableException: Number of sample IDs differs from matrix size!
>>>
$ python
Python 3.5.5 | packaged by conda-forge | (default, Apr  6 2018, 13:43:56)
[GCC 4.2.1 Compatible Apple LLVM 6.1.0 (clang-602.0.53)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import biom
>>> t = biom.Table([], [], ['foo', 'bar'])
>>>

Issue first observed in biocore/unifrac#65 by @gwarmstrong

[wasade]

Fixed in #814