Worker: tweaking lost log events

[josephjclark]

This PR attempts to fix an issue where

I can't reproduce the problem at all, and I don't really have much information to go on from logs.

I've done two things:

• Adjusted where we catch errors in event processing. This may stop the uncaught exception appearing in the logs (it also may not!)
• Add timeout to all messages, based on the server message timeout plus a grace period. This says: give the message N seconds to process, and if it doesn't, go and process another one.

The timeout thing would stop the zombie affect - at least it'll try and process all messages now.

Note that when/if we restore retries, we need to give each retry its own timeout (else retried event will just time themselves out)

Fixes #1178

AI Usage

Please disclose how you've used AI in this work (it's cool, we just want to know!):

• Code generation (copilot but not intellisense)
• Learning or fact checking
• Strategy / design
• Optimisation / refactoring
• Translation / spellchecking / doc gen
• Other
• I have not used AI

You can read more details in our Responsible AI Policy