Harden global revive scheduler execution

[Nergly]

Motivation

• The global scheduled "GlobalReviveUnDeathBan" job could silently miss its configured TimeOfDay run or stop due to uncaught exceptions and async scheduling fragility, making global revives/un-deathbans unreliable.

Description

• Run the scheduler polling task on the main server thread by switching from runTaskTimerAsynchronously to runTaskTimer to avoid async timing fragility.
• Add informational logs at startup, on execution start, and after completion to make scheduled runs and the next run timestamp observable through the plugin logger.
• Wrap the scheduler run() body in a try/catch and log SEVERE on unexpected exceptions to prevent silent task death and aid debugging.
• Changes made in src/com/backtobedrock/augmentedhardcore/runnables/GlobalReviveUnDeathBan.java.

Testing

• Ran the project compile step with mvn -q -DskipTests compile which completed successfully.
• Executed unit tests with mvn -q test during validation and no test failures were reported.

---

Codex Task