MDEV-36868: Inconsistency when shrinking innodb_buffer_pool_size

[dr-m]

• The Jira issue number for this PR is: MDEV-36868

Description

buf_pool_t::resize(): After successfully shrinking the buffer pool, announce the success. The size had already been updated in shrunk(). After failing to shrink the buffer pool, re-enable the adaptive hash index if it had been enabled.

Release Notes

When an attempt to reduce the allocation with SET GLOBAL innodb_buffer_pool_size is made, the adaptive hash index would not always be re-enabled afterwards. Also, no message was output about shrinking the buffer pool.

How can this PR be tested?

SET GLOBAL innodb_buffer_pool_size=16*1024*1024;
SET GLOBAL innodb_buffer_pool_size=8*1024*1024;

and check the server error log.

Basing the PR against the correct MariaDB version

• This is a new feature or a refactoring, and the PR is based against the main branch.
• This is a bug fix, and the PR is based against the earliest maintained branch in which the bug can be reproduced.
  This fixes a regression due to 10.11 MDEV-29445 reorganise innodb buffer pool (and remove buffer pool chunks) #3826.

PR quality check

• I checked the CODING_STANDARDS.md file and my PR conforms to this where appropriate.
• For any trivial modifications to the PR, I am ok with the reviewer making the changes themselves.

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[mariadb-DebarunBanerjee]

The changes look correct to me. Since the impact could be significant let's add a test to protect it.

[mariadb-DebarunBanerjee]

Can you please add a testcase to check it ? Looks like it is doable just checking checking innodb_adaptive_hash_index global which should reflect the correct state.