Make ingestor fault tolerant and production ready

[rjo100]

Right now we haven't done a comprehensive overview of metrics and error handling for the ingestor (and associated pipeline steps like post_process and eventstream). We should:

1. Ensure metric consistency in our new ingestor pipeline with our current processing pipeline
2. Make sure that all new code pathways fail gracefully
   a. Error handling where appropriate
   b. Invalid payloads are rejected with helpful error messages and don't cause unhandled internal errors
3. Set up SLO's for existing event and transaction processing with an eye towards occurrence SLO's

[rjo100]

Need to add more metrics to understand amount of load. Logging is in a decent place

[rjo100]

Need to make SLOs for events and transactions as well
https://www.notion.so/sentry/SLO-Update-Q4-FY23-3a100322fcd945528e4d58e89e339379