[chore][tailsamplingprocessor] Add stability level per metric

[ChrsMark]

Description

open-telemetry/opentelemetry-collector#13756 added support for exposing metrics' stability level in the generated documentation. This PR makes use of this functionality.

We start by setting the stability of all metrics to development. More info about levels can be found at https://github.com/open-telemetry/opentelemetry-specification/blob/v1.49.0/oteps/0232-maturity-of-otel.md#maturity-levels.

Related to:

• Document SemConv compatibility requirements for components stability opentelemetry-collector#11878
• [mdatagen] Consider linking metrics/attributes from metadata.yaml to SemConv definitions opentelemetry-collector#13297
• [receiver/hostmetricsreceiver] semantic conventions transition #35325
• Add stability level per metric #42809

Link to tracking issue

Fixes ~

Testing

~

Documentation

Updated

[ChrsMark]

@portertech please take a look

[edmocosta]

It seems some unit tests are failing due to this change, is that right?

Running target 'test' in module 'processor/tailsamplingprocessor' as part of group 'processor-1'
make --no-print-directory -C processor/tailsamplingprocessor test

=== FAIL: . TestMetricsWithComponentID (0.00s)
    processor_telemetry_test.go:316: [Description not equal:
        expected: Count of traces that were sampled or not per sampling policy
        actual: Count of traces that were sampled or not per sampling policy [development]]
    processor_telemetry_test.go:316: [Description not equal:
        expected: Latency (in microseconds) of a given sampling policy
        actual: Latency (in microseconds) of a given sampling policy [development]]

=== FAIL: . TestMetricsCountSampled/not_sampled/otelcol_processor_tail_sampling_global_count_traces_sampled (0.00s)
    processor_telemetry_test.go:598: [Description not equal:
        expected: Global count of traces that were sampled or not by at least one policy
        actual: Global count of traces that were sampled or not by at least one policy [development]]

...

[ChrsMark]

It seems some unit tests are failing due to this change, is that right?

Running target 'test' in module 'processor/tailsamplingprocessor' as part of group 'processor-1'
make --no-print-directory -C processor/tailsamplingprocessor test

=== FAIL: . TestMetricsWithComponentID (0.00s)
    processor_telemetry_test.go:316: [Description not equal:
        expected: Count of traces that were sampled or not per sampling policy
        actual: Count of traces that were sampled or not per sampling policy [development]]
    processor_telemetry_test.go:316: [Description not equal:
        expected: Latency (in microseconds) of a given sampling policy
        actual: Latency (in microseconds) of a given sampling policy [development]]

=== FAIL: . TestMetricsCountSampled/not_sampled/otelcol_processor_tail_sampling_global_count_traces_sampled (0.00s)
    processor_telemetry_test.go:598: [Description not equal:
        expected: Global count of traces that were sampled or not by at least one policy
        actual: Global count of traces that were sampled or not by at least one policy [development]]

...

Ouch, missed that believing it was just flakiness based on the fact that the change should only affect auto-generated code. Thank's for catching this! Should be fixed now.

[lukaszgryglicki]

/easycla