Perf: Optimize performance of ByteViewGroupValueBuilder on batches with inlined views

[zhuqi-lucas]

Which issue does this PR close?

Optimize following cases, and add more fast path.

do_append_val_inner
do_equal_to_inner

This is wasteful if there is no data buffer, in which case we can improve the performance.

Also, i do some fast path which is done in arrow-rs.

• Closes #16330

Rationale for this change

Optimize following cases, and add more fast path.

do_append_val_inner
do_equal_to_inner

What changes are included in this PR?

Optimize following cases, and add more fast path.

do_append_val_inner
do_equal_to_inner

Are these changes tested?

Yes

Are there any user-facing changes?

No

[Rachelint]

Thanks @zhuqi-lucas ! How about the benchmark result?

[zhuqi-lucas]

Thank you @Rachelint , it seems no performance improvement for clickbench from my local Mac benchmark, i need to investigate further.

[Rachelint]

Thank you @Rachelint , it seems no performance improvement for clickbench from my local Mac benchmark, i need to investigate further.

Yes... I made a simple try only in append path before, and also found no obvious improvement... We indeed need further investigation maybe

[zhuqi-lucas]

Thank you @Rachelint , it seems no performance improvement for clickbench from my local Mac benchmark, i need to investigate further.

Yes... I made a simple try only in append path before, and also found no obvious improvement... We indeed need further investigation maybe

I agree, converted it a draft now.

[github-actions]

Thank you for your contribution. Unfortunately, this pull request is stale because it has been open 60 days with no activity. Please remove the stale label or comment or this will be closed in 7 days.