[mdatagen] Add semconv reference for metrics in metadata schema

[ChrsMark]

Description

This PR adds support in mdatagen for defining SemConv reference for metrics.

Enhancing your metrics/signals in metadata.yaml like bellow:

metrics:
  system.cpu.time:
    enabled: true
    stability:
      level: beta
    description: Monotonic cumulative sum int metric enabled by default.
    unit: s
    sum:
      value_type: int
      monotonic: true
      aggregation_temporality: cumulative
    semantic_convention:
      semconv_ref: https://github.com/open-telemetry/semantic-conventions/blob/v1.37.0/docs/system/system-metrics.md#metric-systemcputime

Will provide the following generated docs:

system.cpu.time

Monotonic cumulative sum int metric enabled by default.

Unit	Metric Type	Value Type	Aggregation Temporality	Monotonic	Stability	Semantic Convention
s	Sum	Int	Cumulative	true	beta	system.cpu.time

Validation is added so as to ensure that provided SemConv urls match the top_level semconv version as well as the metric's name.

Link to tracking issue

Related to #13910 and #13297.

Testing

Tuned

Documentation

Tuned

[codecov]

Codecov Report

❌ Patch coverage is 94.05941% with 6 lines in your changes missing coverage. Please review.
✅ Project coverage is 92.27%. Comparing base (6f29b34) to head (2442bf8).

Files with missing lines	Patch %	Lines
cmd/mdatagen/internal/metric.go	77.77%	3 Missing and 3 partials ⚠️

❌ Your patch check has failed because the patch coverage (94.05%) is below the target coverage (95.00%). You can increase the patch coverage or adjust the target coverage.

Additional details and impacted files

@@           Coverage Diff           @@
##             main   #13920   +/-   ##
=======================================
  Coverage   92.27%   92.27%           
=======================================
  Files         657      657           
  Lines       41111    41207   +96     
=======================================
+ Hits        37936    38025   +89     
- Misses       2173     2177    +4     
- Partials     1002     1005    +3     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[codeboten]

i like this change a lot. would be great to get this in

[ChrsMark]

i like this change a lot. would be great to get this in

I'll get back into finalising this and opening for review soonish!

[ChrsMark]

@codeboten @mx-psi I've opened this for review now. Feel free to take a look when you get the chance, thank's!

[mx-psi]

I think the semconv_ prefix here is redundant since this is already in the semantic_convention section.

[mx-psi]

I am not sure what this field is, but it seems like we should either drop this or make sure that sem_conv_version and the version linked in semconv_ref are aligned somehow

[ChrsMark]

This is sth that already exists and links the whole component to a specific version of SemConv. It's not really used apart from a few places inside the templates that mdatagen populates i.e. logs.go.tmpl and metrics.go.tmpl.

The PR comes with a pattern checking that also verifies if the provided reference's version matches with this generic one: https://github.com/open-telemetry/opentelemetry-collector/pull/13920/files#diff-b33d4fa1a4595583d310c0a381ce3849aefb42419451584fd0e1c43bd16f5263R124-R128. (also covered by this test)

[mx-psi]

Okay, sorry, another thing I missed. I think that makes sense, though maybe we can remove it entirely if it's not widely used?

I am worried that we won't be able to support transition scenarios where we want to target two semconv versions