Add __repr__ for Column and ColumnAccessor

[shwina]

Summary:

• Add a __repr__ for Column (thin wrapper around the __repr__ of the underlying pa.Array)
• Add a __repr__ for ColumnAccessor (similar to pa.Table, shows the names/types of the columns of the ColumnAccessor)

Additional info:

Debugging is sometimes made painful by the fact that we don't have a __repr__ for columns and column accessors. For example, here's what a ColumnAccessor and a Column currently look like when printed...:

In [2]: cudf.DataFrame({'a': [1, 2, 3], "b": [4, 5, 6], "z_1": [2, 3, 4]})._data
Out[2]: ColumnAccessor(OrderedColumnDict([('a', <cudf.core.column.numerical.NumericalColumn object at 0x7f0306336f80>), ('b', <cudf.core.column.numerical.NumericalColumn object at 0x7f03062a05f0>), ('z_1', <cudf.core.column.numerical.NumericalColumn object at 0x7f03062a0e60>)]), multiindex=False, level_names=(None,))

In [3]: cudf.Series([1, 2, None, 3])._column
Out[3]: <cudf.core.column.numerical.NumericalColumn at 0x7f2190746710>

After this PR:

In [2]: cudf.DataFrame({'a': [1, 2, 3], "b": [4, 5, 6], "z_1": [2, 3, 4]})._data
Out[2]:
ColumnAccessor(multiindex=False, level_names=(None,))
a: int64
b: int64
z_1: int64

In [3]: cudf.Series([1, 2, None, 3])._column
Out[3]:
<cudf.core.column.numerical.NumericalColumn object at 0x7f3e90c2ac20>
[
  1,
  2,
  null,
  3
]
dtype: int64

[shwina]

Should ColumnAccessor print more like a dictionary to reinforce that it's a mapping?

[isVoid]

Does {'a': 1, 'b': 2} imply the keys are unordered in python's context? I suggest using extra symbols to also indicate the orderness.

[shwina]

@isVoid currently we don't print ColumnAccessor like dicts -- so there's no implication of unordered-ness.

[shwina]

rerun tests

[codecov]

Codecov Report

Merging #7531 (0afb9a2) into branch-0.19 (7871e7a) will increase coverage by 0.51%.
The diff coverage is 92.00%.

@@               Coverage Diff               @@
##           branch-0.19    #7531      +/-   ##
===============================================
+ Coverage        81.86%   82.38%   +0.51%     
===============================================
  Files              101      101              
  Lines            16884    17340     +456     
===============================================
+ Hits             13822    14285     +463     
+ Misses            3062     3055       -7     

Impacted Files	Coverage Δ
python/cudf/cudf/core/column/column.py	87.83% <83.33%> (+0.07%)	⬆️
python/cudf/cudf/core/column/decimal.py	93.33% <90.90%> (-1.54%)	⬇️
python/cudf/cudf/core/column/string.py	86.76% <100.00%> (+0.26%)	⬆️
python/cudf/cudf/core/column_accessor.py	95.45% <100.00%> (+0.14%)	⬆️
python/cudf/cudf/utils/cudautils.py	52.94% <100.00%> (+2.55%)	⬆️
python/cudf/cudf/utils/gpu_utils.py	53.65% <0.00%> (-4.88%)	⬇️
python/cudf/cudf/core/abc.py	87.23% <0.00%> (-1.14%)	⬇️
python/cudf/cudf/core/column/numerical.py	94.85% <0.00%> (-0.17%)	⬇️
python/cudf/cudf/io/feather.py	100.00% <0.00%> (ø)
... and 48 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 9618a81...0afb9a2. Read the comment docs.

[kkraus14]

@gpucibot merge