feat: add server metrics

[matthisholleville]

Closes #

📑 Description

This commit add metrics path & the analyzer error metrics in the codebase.

The changes have been made across all analyzers and include the addition of a new metric with label values for the analyzer's name, analyzed object's name, and namespace.

The metric's value is set to the length of the analyzer objects failures.

✅ Checks

• My pull request adheres to the code style of this project
• My code requires changes to the documentation
• I have updated the documentation as required
• All the tests have passed

ℹ Additional Information

$ curl http://localhost:8080/metrics
# HELP analyzer_errors Number of errors detected by analyzer
# TYPE analyzer_errors gauge
analyzer_errors{analyzer_name="HorizontalPodAutoscaler",namespace="k8sgpt",object_name="fake-hpa"} 1
analyzer_errors{analyzer_name="Ingress",namespace="k8sgpt",object_name="fake-ingress"} 3
...

[thschue]

Good work, I think the Metric should be a Counter instead of a Gauge, but the rest look good!

[matthisholleville]

Thank you for your feedback @thschue. From my understanding, the metrics can increase or decrease depending on the number of issues. The user can partially fix errors and the metrics will decrease accordingly.

On the other hand, the counter can only increase which is not our case.

What do you think?

[thschue]

Thank you for your feedback @thschue. From my understanding, the metrics can increase or decrease depending on the number of issues. The user can partially fix errors and the metrics will decrease accordingly.

On the other hand, the counter can only increase which is not our case.

What do you think?

Sorry, had the wrong use-case in mind. The gauge is ok in this case!

[thschue]

lgtm