trace: fix span leak in batchSpanProcessor

[korniltsev]

Description

When a span ReadOnlySpan is processed and then exported by batchSpanProcessor it may be not garbage collected. A strong pointer to a ReadOnlySpan may be kept in batchSpanProcessor.batch slice.

Environment

• OS: Darwin
• Architecture: arm64
• Go Version: go version devel go1.23-eaa7d9ff86 Mon Jun 3 14:56:37 2024 +0000 darwin/arm64
• opentelemetry-go version: f80064a

Steps To Reproduce

	bsp := sdktrace.NewBatchSpanProcessor(tracetest.NewInMemoryExporter())
	tp := sdktrace.NewTracerProvider(
		sdktrace.WithSpanProcessor(bsp),
	)

	_, span := tp.Tracer("tracer").Start(ctx, "leaked_span")
	span.End()

	tp.ForceFlush(ctx)

Expected behavior

There are no references to the leaked_span in the batchSpanProcessor

Actual behavior

There is a references to the leaked_span in the batchSpanProcessor.batch

PR description

In this PR batchSpanProcessor clears references to spans after export.

[linux-foundation-easycla]

The committers listed above are authorized under a signed CLA.

• ✅ login: korniltsev / name: Tolya Korniltsev (e22d776)

[dashpole]

Would you mind either updating the description or opening an issue with the problem you encountered? In which scenarios are we leaking spans, and how did you encounter the issue?

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 84.8%. Comparing base (478f85b) to head (e22d776).
Report is 2 commits behind head on main.

Additional details and impacted files

@@          Coverage Diff          @@
##            main   #5518   +/-   ##
=====================================
  Coverage   84.8%   84.8%           
=====================================
  Files        269     269           
  Lines      18079   18080    +1     
=====================================
+ Hits       15333   15336    +3     
+ Misses      2420    2418    -2     
  Partials     326     326           

Files	Coverage Δ
sdk/trace/batch_span_processor.go	79.0% <100.0%> (+0.1%)	⬆️

... and 2 files with indirect coverage changes

[korniltsev]

@dashpole Updated description of the PR