[C++] Clarify ChunkedArray chunking strategy and policy

[asfimport]

See discussion on ARROW-6784 and #5686. Among the questions:

• Do Arrow users control the chunking, or is it an internal implementation detail they should not manage?
• If users control it, how do they control it? E.g. if I call Take and use a ChunkedArray for the indices to take, does the chunking follow how the indices are chunked? Or should we attempt to preserve the mapping of data to their chunks in the input table/chunked array?
• If it's an implementation detail, what is the optimal chunk size? And when is it worth reshaping (concatenating, slicing) input data to attain this optimal size? 

Reporter: Neal Richardson / @nealrichardson
Assignee: Wes McKinney / @wesm

Related issues:

• [C++] Split arrow::ChunkedArray into arrow/chunked_array.h (relates to)

PRs and other links:

• GitHub Pull Request #7510

_{Note: This issue was originally created as ARROW-7012. Please see the migration documentation for further details.}

[asfimport]

Wes McKinney / @wesm:
In general, this is not something that users should be too concerned with. The new kernels framework provides a configurability knob (ExecContext::exec_chunksize) for selecting the upper limit for the size of chunks that are processed

[asfimport]

Wes McKinney / @wesm:
I think we can close this. I don't think that users need to be especially concerned with this, though the chunksize will be publicly configurable for expression execution

[asfimport]

Neal Richardson / @nealrichardson:
We still may want to document that fact

[asfimport]

Wes McKinney / @wesm:
Agreed. Where do you think this should go, in the docstrings for ChunkedArray, in arrow/compute/README.md, somewhere else?

[asfimport]

Neal Richardson / @nealrichardson:
IMO it should go in the main ChunkedArray docs where the class is described/explained. There are multiple ways you could end up at the question of how you should chunk arrays, so I think compute is not quite the right place. Either way, we can move it/link to it from both/all relevant places.

[asfimport]

Wes McKinney / @wesm:
OK. I'll take care of this.

[asfimport]

Antoine Pitrou / @pitrou:
Issue resolved by pull request 7510
#7510