Fix Redis fallback ack_deadline_ms handling by updating message IDs …

[Soroushsrd]

Problem

The Redis fallback implementation has an issue where message timeout tracking is based on a message's creation time rather than when it was last received. This results in:

• Messages possibly being re-queued too soon
• Duplicate processing of messages
• Inconsistent ack deadlines

Solution

This PR fixes the issue by:

1. Breaking down the atomic BRPOPLPUSH operation into separate BRPOP and LPUSH operations
2. Generating a new ID with the current timestamp when moving messages to the processing queue
3. Ensuring the ack deadline is measured from when a message is received, not created

Fixes #102

[Soroushsrd]

Fixed the linting issue and amended the earlier commit

[svix-jplatte]

Hey, looks like tests are failing with your change. You can test yourself by running sudo docker compose -f testing-docker-compose.yml up (with -d if you want it to keep running in the background), then cargo test redis_fallback.