DAOS-17550 object: refine dc_array_set_size processing

[liuxuezhao]

1. use memcmp rather than strncmp in merge_key strncmp can cause some keys be incorrectly ignored
2. refine list dkey by providing more memory to save RPC round-trip.

Steps for the author:

• [ ] Commit message follows the guidelines.
• [ ] Appropriate Features or Test-tag pragmas were used.
• [ ] Appropriate Functional Test Stages were run.
• [ ] At least two positive code reviews including at least one code owner from each category referenced in the PR.
• [ ] Testing is complete. If necessary, forced-landing label added and a reason added in a comment.

After all prior steps are complete:

• [ ] Gatekeeper requested (daos-gatekeeper added as a reviewer).

[github-actions[bot]]

Ticket title is 'truncating an EC file to 0 does not work on the first try' Status is 'In Review' Labels: 'scrubbed_2.8' Job should run at elevated priority (1) https://daosio.atlassian.net/browse/DAOS-17550

[mchaarawi]

i have not tested on aurora yet, but will do soon

[liuxuezhao]

im going to revert my +1 because the performance of the patch on aurora increases by 5x.

i just tested the memcmp fix, and that works as expected. but the "improvement" in the array code made truncate worse my 5x: With 4:26.38 minutes Without 0:50.03 minutes

let's remove the other changes from the PR and just keep the memcmp fix please

could you please test again with the refreshed PR? the original code only list 5 dkeys every time it looks not very good, you may tune the number of "ENUM_DESC_NR_MAX" and the nr/buf_size in sub_anchors_prep() to see what is better on aurora. thanks

[daosbuild3]

Test stage Functional Hardware Medium Verbs Provider completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-16404/2/execution/node/1372/log