2.4.0: Change interval to prevent scrape timeouts

Prometheus and Victoriametrics limits the scrape timeout
to be equal or less than scrape interval. Reducing the
defaults to 55 should prevent scrape timeouts. When using
statsd use --interval 60 to ensure the entire period is
covered.

Ref: https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1281

2.3.0: Only Normalize Task CPU limits

Container limits are always in CPU shares, while task limits are in
vCpus

2.2.0: Handle stats for STOPPED containers

Reported in this PR: https://github.com/Spaced-Out/ecs-container-exporter/pull/1

2.0.1

Fix crash due to a typo

2.0.0

Bump version and update Readme