Add support for monitoring pipelines and alerting

[VJftw]

Enhancement

Current behaviour

Lack of monitoring and alerting

Disadvantages of current behaviour

• It is hard to know when a pipeline has failed or how often a pipeline is failing

New behaviour

Implement monitoring/alerting around pipelines (maybe via prometheus metrics)

Advantages of new behaviour

• Users can be alerted when their pipeline fails or has a change in the trend of failures