[FLINK-37470] Improve JobManager Deployment / Pod error handling

[gyfora]

What is the purpose of the change

The operator correctly reacts to certain type of pod errors such as image pull/crash loop when they happen on main / sidecar containers.

However these errors are ignored on initcontainers. Furthermore, volume mount errors which do not manifest in clear container error statueses are not recorded at all.

This PR aims to collectively fix this and greatly improve startup error handling

Brief change log

• Check and catch all container and init container error statuses
• Add special logic to handle common volume mount errors based on pod events in case of failed startup

Verifying this change

New unit tests + manually verified on different local and prod envs

Does this pull request potentially affect one of the following parts:

• Dependencies (does it add or upgrade a dependency): no
• The public API, i.e., is any changes to the CustomResourceDescriptors: no
• Core observer or reconciler logic that is regularly executed: yes

Documentation

• Does this pull request introduce a new feature? no