GH-44393: [C++][Compute] Swizzle vector functions

[zanmato1984]

Rationale for this change

For background please see #44393.

When implementing the "scatter" function requested in #44393, I found it also useful to make it a public vector API. After a painful thinking, I decided to name it "permute". And when implementing permute, I found it fairly easy to implement it by first computing the "reverse indices" of the positions, and then invoking the existing "take", where I think "reverse_indices" itself can also be a useful public vector API. Thus the PR categorized them as "placement functions".

What changes are included in this PR?

Implement placement API reverse_indices and permute, where permute(values, indices) is implemented as take(values, reverse_indices(indices)).

I also put a small UT Permute.IfElse demonstrating how permute can serve as a building block of implementing special forms.

Are these changes tested?

UT included.

Are there any user-facing changes?

Yes, new public APIs added. Documents updated.

• GitHub Issue: [C++][Compute] "Scatter" vector functions #44393

[github-actions]

⚠️ GitHub issue #44393 has been automatically assigned in GitHub to PR creator.

[zanmato1984]

Hi @felipecrv , the renaming is done. Would you like to proceed with the review? Thank you.

[zanmato1984]

Hi @pitrou @felipecrv @mapleFU , shall we move on with this? I'll need this for my new (and possibly formal) implementation of special form.

Also cc @bkietz .

Thanks.

[pitrou]

Thank you @zanmato1984 and sorry for the delay in reviewing. You'll find some comments below.

Also, can you update this PR with recent git main and fix any conflicts?

[pitrou]

Would it clearer to name it something like output_length? (you would have to offset its definition by 1)
@felipecrv

[zanmato1984]

Hmm, it was exactly output_length until I modified it according to @felipecrv 's suggestion. See #44394 (comment)

[zanmato1984]

I personally prefer output_length as it's more straightforward and saves a little implementation code actually. On the other hand, max_index contains more "semantic" though.

[pitrou]

Ahah. Then perhaps we can just update the doc to say that the output length will be max_index + 1 if this option is set?

[zanmato1984]

Sure. Is updating here 53f3c33 (#44394) enough?

[zanmato1984]

Hi @pitrou , you will see my commits and comments one by one, this is me addressing each of your comments. I'll explicitly let you know once they are all done. Before that, you can just ignore the notification.

Thank you for the review, appreciate that!

[zanmato1984]

Hi @pitrou , I think I've addressed all the comments before. Would you take a second look? Thanks.