Add prestop hook

[WalBeh]

Summary of changes

For rolling restarts of the crateDB pods, we want to notify cratedb with an alter cluster decommission¹ to allow proper handling of running queries and shard handling:

• the preStop Lifecycle² executes dc_util, which is provided via https://cdn.crate.io/downloads/dc_util_x86_64
• dc_util sets the crateDB decommission timeout to 720s and sets decomission force to true to force decommission in cases crateDB would roll-back the decomission.
• termination_grace_period_seconds on the POD is set to 900s to allow the decommission to finish.

In case the statefulset is scaled to 0 (scale down, or manual triggered by an administrator), dc_util tries to detect that and does NOT trigger an alter cluster decommission and POD Termination is triggered by kubelet sending a SIGTERM.

preStop seems to behave some what fail-save. Eg. if dc_util is not available or cannot be executed, kubelet continues by sending SIGTERM.

Checklist

• Link to issue this PR refers to: https://github.com/crate/cloud/issues/2127
• Relevant changes are reflected in CHANGES.rst
• Added or changed code is covered by tests
• Documentation has been updated if necessary
• Changed code does not contain any breaking changes (or this is a major version change)

Footnotes

1. https://cratedb.com/docs/crate/reference/en/latest/sql/statements/alter-cluster.html#decommission-nodeid-nodename ↩

2. https://kubernetes.io/docs/concepts/workloads/pods/pod-lifecycle/ ↩