fix(otel): Missing addLink method and Fiber handling

[PROFeNoM]

Description

• SpanInterface::addLink was added with Add Span::addLink() and Meter::createGauge() open-telemetry/opentelemetry-php#1289 and is included in 1.1+. It is currently missing from our shim. For backward compatibility reasons, ImmutableSpan will take a new argument, totalRecordedLinks if the latter method exists.
• Improve fiber bound context storage open-telemetry/opentelemetry-php#1270 modified Fiber bound context storage, and hence required changes to Context::storage. For backward compatibility reasons, we have to keep both initializations.
• To proactively spot such changes and ensure backward compatibility, the test_opentelemetry_1 suite (currently: @stable) will be split into two: @beta and 1.0.*.
• There are some white space changes in the PR 😬 Sorry, I didn't manage to remove them :-(
• I just changed the Fiber test back to .phpt; it was driving me crazy and I couldn't otherwise properly reset the context between tests (...I didn't have time to look further into it). This way, it's appropriately reset :-)

Reviewer checklist

• Test coverage seems ok.
• Appropriate labels assigned.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 57.14286% with 21 lines in your changes missing coverage. Please review.

Project coverage is 80.94%. Comparing base (31382ec) to head (e44f4e4).

Files with missing lines	Patch %	Lines
src/DDTrace/OpenTelemetry/Span.php	59.09%	18 Missing ⚠️
src/DDTrace/OpenTelemetry/Context.php	40.00%	3 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##             master    #2849      +/-   ##
============================================
- Coverage     81.05%   80.94%   -0.11%     
- Complexity     2517     2526       +9     
============================================
  Files           146      146              
  Lines         14654    14680      +26     
============================================
+ Hits          11878    11883       +5     
- Misses         2776     2797      +21     

Flag	Coverage Δ
tracer-extension	78.18% <ø> (ø)
tracer-php	82.10% <57.14%> (-0.16%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
src/DDTrace/OpenTelemetry/Context.php	82.81% <40.00%> (-3.48%)	⬇️
src/DDTrace/OpenTelemetry/Span.php	85.42% <59.09%> (-6.58%)	⬇️

... and 1 file with indirect coverage changes

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 31382ec...e44f4e4. Read the comment docs.

[pr-commenter]

Benchmarks

Benchmark execution time: 2024-09-16 08:08:18

Comparing candidate commit e4b186a in PR branch alex/fix/otel-addlink-fiber with baseline commit 1a7182c in branch master.

Found 3 performance improvements and 5 performance regressions! Performance is the same for 170 metrics, 0 unstable metrics.

scenario:LaravelBench/benchLaravelBaseline-opcache

• 🟥 execution_time [+86.391µs; +206.589µs] or [+2.856%; +6.830%]

scenario:PDOBench/benchPDOBaseline

• 🟩 execution_time [-13.842µs; -11.256µs] or [-7.187%; -5.844%]

scenario:PDOBench/benchPDOOverheadWithDBM

• 🟩 execution_time [-14.699µs; -12.174µs] or [-4.740%; -3.926%]

scenario:SamplingRuleMatchingBench/benchRegexMatching2

• 🟥 execution_time [+333.793ns; +713.207ns] or [+4.872%; +10.410%]

scenario:SamplingRuleMatchingBench/benchRegexMatching3

• 🟥 execution_time [+149.073ns; +441.727ns] or [+2.140%; +6.340%]

scenario:SamplingRuleMatchingBench/benchRegexMatching4-opcache

• 🟥 execution_time [+142.816ns; +503.984ns] or [+2.102%; +7.419%]

scenario:TraceSerializationBench/benchSerializeTrace

• 🟥 execution_time [+3.996µs; +7.604µs] or [+2.374%; +4.518%]

scenario:WordPressBench/benchWordPressOverhead

• 🟩 execution_time [-21.847ms; -21.541ms] or [-83.051%; -81.888%]

[PROFeNoM]

I'm putting it back to draft. +7-9% performance regression for a simple check is simply too much. I'll try and optimize that a little bit.

[bwoebi]

@PROFeNoM Feels like the benchmarks are simply flaky. ... But I suppose for that one the micro-optimization is worth it.

[pr-commenter]

Benchmarks [ tracer ]

Benchmark execution time: 2024-09-16 13:24:15

Comparing candidate commit e44f4e4 in PR branch alex/fix/otel-addlink-fiber with baseline commit 31382ec in branch master.

Found 2 performance improvements and 4 performance regressions! Performance is the same for 172 metrics, 0 unstable metrics.

scenario:ContextPropagationBench/benchInject64Bit

• 🟥 execution_time [+188.236ns; +617.764ns] or [+2.531%; +8.307%]

scenario:ContextPropagationBench/benchInject64Bit-opcache

• 🟥 execution_time [+183.937ns; +568.063ns] or [+2.446%; +7.554%]

scenario:LogsInjectionBench/benchLogsInfoInjection-opcache

• 🟩 execution_time [-439.363ns; -243.637ns] or [-5.657%; -3.137%]

scenario:PDOBench/benchPDOBaseline-opcache

• 🟥 execution_time [+9.677µs; +14.422µs] or [+5.432%; +8.096%]

scenario:PDOBench/benchPDOOverhead-opcache

• 🟥 execution_time [+14.523µs; +17.533µs] or [+5.064%; +6.114%]

scenario:WordPressBench/benchWordPressOverhead

• 🟩 execution_time [-21.839ms; -21.599ms] or [-82.871%; -81.961%]

[bwoebi]

Thanks, looks good :-)