feat: introduce metrics for failed evaluations

[odubajDT]

This PR

• introduced a new metric which buckets evaluations by reason
• small refactoring
• unit tests

Related Issues

Fixes #559

How to test

Run flagd via

make run

Run

curl -X POST "localhost:8013/schema.v1.Service/ResolveString" -d '{"flagKey":"fibAlgo","context":{}}' -H "Content-Type: application/json"
curl -X POST "localhost:8013/schema.v1.Service/ResolveString" -d '{"flagKey":"myStringFlag","context":{}}' -H "Content-Type: application/json"
curl -X POST "localhost:8013/schema.v1.Service/ResolveString" -d '{"flagKey":"myStringFlagNonExisting","context":{}}' -H "Content-Type: application/json"
curl -X POST "localhost:8013/schema.v1.Service/ResolveString" -d '{"flagKey":"myStringFlagNonExisting2","context":{}}' -H "Content-Type: application/json"
curl -X POST "localhost:8013/schema.v1.Service/ResolveInt" -d '{"flagKey":"myStringFlag","context":{}}' -H "Content-Type: application/json"

Open http://localhost:8014/metrics and see something like the following

# HELP impressions_total The number of evaluation for a given flag
# TYPE impressions_total counter
impressions_total{feature_flag_key="fibAlgo",feature_flag_provider_name="flagd",feature_flag_reason="DEFAULT",feature_flag_variant="recursive",otel_scope_name="openfeature/flagd",otel_scope_version=""} 1
impressions_total{feature_flag_key="myStringFlag",feature_flag_provider_name="flagd",feature_flag_reason="STATIC",feature_flag_variant="key1",otel_scope_name="openfeature/flagd",otel_scope_version=""} 1

and

# HELP reasons_total The number of evaluations for a given reason
# TYPE reasons_total counter
reasons_total{feature_flag_provider_name="flagd",feature_flag_reason="DEFAULT",otel_scope_name="openfeature/flagd",otel_scope_version=""} 1
reasons_total{feature_flag_provider_name="flagd",feature_flag_reason="STATIC",otel_scope_name="openfeature/flagd",otel_scope_version=""} 1
reasons_total{exception_type="FLAG_NOT_FOUND",feature_flag_provider_name="flagd",feature_flag_reason="ERROR",otel_scope_name="openfeature/flagd",otel_scope_version=""} 2
reasons_total{exception_type="TYPE_MISMATCH",feature_flag_provider_name="flagd",feature_flag_reason="ERROR",otel_scope_name="openfeature/flagd",otel_scope_version=""} 1

[codecov]

Codecov Report

Merging #584 (17b1eab) into main (c25d0ef) will increase coverage by 0.27%.
The diff coverage is 88.09%.

@@            Coverage Diff             @@
##             main     #584      +/-   ##
==========================================
+ Coverage   70.78%   71.05%   +0.27%     
==========================================
  Files          22       22              
  Lines        2249     2277      +28     
==========================================
+ Hits         1592     1618      +26     
- Misses        592      594       +2     
  Partials       65       65              

Impacted Files	Coverage Δ
core/pkg/eval/json_evaluator.go	87.72% <33.33%> (ø)
core/pkg/service/flag-evaluation/flag_evaluator.go	67.82% <70.00%> (-0.52%)	⬇️
core/pkg/eval/ievaluator.go	100.00% <100.00%> (ø)
core/pkg/telemetry/metrics.go	100.00% <100.00%> (ø)

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[thisthat]

In the ticket, the agreement was to create a new metric instead of reusing impressions.

[beeme1mr]

This works great from a functionality perspective. I'll let others comment on the code itself.