Add null check to StrictContextStorage

[laurit]

https://ge.opentelemetry.io/s/y3ppcyqy2q5v2/tests/task/:instrumentation:reactor:reactor-netty:reactor-netty-1.0:javaagent:test/details/io.opentelemetry.javaagent.instrumentation.reactornetty.v1_0.ReactorNettyHttpClientUsingFromTest/shouldNotLeakConnections()?expanded-stacktrace=WyIwIl0&top-execution=1
Also see open-telemetry/opentelemetry-java-instrumentation#7304
Apparently ReferenceQueue#remove() can return null on openj9.

[codecov]

Codecov Report

Attention: 1 lines in your changes are missing coverage. Please review.

Files	Coverage Δ
...io/opentelemetry/context/StrictContextStorage.java	81.65% <50.00%> (-0.76%)	⬇️

... and 2 files with indirect coverage changes

📢 Thoughts on this report? Let us know!.

[jkwatson]

Can you provide a unit test for this change?

[laurit]

@jkwatson I believe that this only happens on openj9. To create a test that reproduces we would first need to run the tests with openj9. Secondly looking at https://github.com/eclipse-openj9/openj9/blob/78038a17db84716d97cb57e1b76bec7975a00bf7/jcl/src/java.base/share/classes/java/lang/ref/ReferenceQueue.java#L139-L162 I'd guess for it to return null there would have to be a spurious wakeup from wait or perhaps multiple threads calling remove. I doubt that it is possible to write an unit test that would reliably reproduce this with a single run.

[trask]

maybe worth adding ^^ as a code comment?