Auto-restart endpoints with kernel bugs

[olegbbtr]

Problem description / Motivation

Right now, we have to manually restart them.

Implementation ideas

Should we do it for any page fault, or for specific messages we are aware about.

[sharnoff]

See also discussion here: https://neondb.slack.com/archives/C0807C9SSJ2/p1748615734988929

[stradig]

We are not 100% sure if we want to do this or have control plane do the restart. Let's continue discussing offline and decide them.

[zenithdb-bot-dev]

This issue was moved to Jira: LKB-1538