reset BA destroy flag on BA create

[bratpiorka]

This PR ensures the global base allocator’s destroyed flag is reset on creation and re-enables the test_init_teardown in CI.

• Reset ba_is_destroyed to false in umf_ba_create_global
• Add and adjust debug logs in the memory tracker and base allocator
• Remove exclusion of test_init_teardown from CI workflows

[Copilot]

Pull Request Overview

This PR ensures the global base allocator’s destroyed flag is reset on creation and re-enables the test_init_teardown in CI.

• Reset ba_is_destroyed to false in umf_ba_create_global
• Add and adjust debug logs in the memory tracker and base allocator
• Remove exclusion of test_init_teardown from CI workflows

Reviewed Changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated no comments.

File	Description
src/provider/provider_tracking.c	Adjusted LOG_ERR formatting and added a LOG_DEBUG inside the tracker-empty check
src/base_alloc/base_alloc_global.c	Added debug on destroy, updated create log, and reset destroy flag
.github/workflows/reusable_multi_numa.yml	Removed test_init_teardown exclusion for RHEL/SLES tests
.github/workflows/nightly.yml	Removed test_init_teardown exclusion from nightly test runs

Comments suppressed due to low confidence (3)

.github/workflows/reusable_multi_numa.yml:59

• [nitpick] Since test_init_teardown is now enabled, this TODO about excluding it is outdated and can be removed to avoid confusion.

# TODO: add issue for -E test_init_teardown - it is not clear why it fails

src/base_alloc/base_alloc_global.c:77

• [nitpick] The log message was changed from 'UMF base allocator created' to 'global base allocator created', which may reduce consistency with other UMF logs—consider restoring or clarifying the prefix for uniformity.

LOG_DEBUG("global base allocator created");

src/provider/provider_tracking.c:772

• [nitpick] This debug log runs inside a loop for every abandoned allocation, which may flood logs or impact performance—consider using a lower-verbosity level (e.g., TRACE) or batching these messages.

LOG_DEBUG(

[lukaszstolarczuk]

LGTM