Add assertion support for Span hasException(null)

[trask]

When hasException(null) is called, it now asserts that the span has no exception event instead of throwing a NullPointerException.

This would be convenient in open-telemetry/opentelemetry-java-instrumentation#16049, e.g. when chaining:

     ...
     .hasException(emitExceptionsAsLogs()? null : exception)
     ...

I think the behavior reasonably follows convention:

• hasSchemaUrl(@Nullable String) asserts the schema URL equals the given value (including null)
• hasBody(@Nullable Value<?>) asserts the body equals the given value (including null)
• equalTo(key, null) in hasAttributesSatisfyingExactly asserts the attribute value is null/absent

This last one we use heavily in instrumentation repo, e.g.

   hasAttributesSatisfyingExactly(
     ...
     equalTo(DB_OPERATION, emitStableDatabaseSemconv() ? null : "SELECT")
     ...
   )

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 90.16%. Comparing base (fd7d014) to head (0c3a9ab).
⚠️ Report is 4 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff            @@
##               main    #8033   +/-   ##
=========================================
  Coverage     90.16%   90.16%           
- Complexity     7483     7485    +2     
=========================================
  Files           836      836           
  Lines         22562    22567    +5     
  Branches       2237     2239    +2     
=========================================
+ Hits          20343    20348    +5     
  Misses         1515     1515           
  Partials        704      704           

☔ 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.

[jack-berg]

This is another case of: #7359 (review)

I support retaining the ability to add the @Nullable annotation to parameters, and think its useful here. Also, there was never any followup to #7359 to update VERSIONING.md to disallow this type of change, so it's still allowed.