Remove net5.0 from Zipkin and replace with net6.0

[cijothomas]

Part of #3147

[codecov]

Codecov Report

Merging #3222 (f89f0f7) into main (11699a3) will decrease coverage by 0.29%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##             main    #3222      +/-   ##
==========================================
- Coverage   85.74%   85.44%   -0.30%     
==========================================
  Files         260      260              
  Lines        9366     9366              
==========================================
- Hits         8031     8003      -28     
- Misses       1335     1363      +28     

Impacted Files	Coverage Δ
...rc/OpenTelemetry.Exporter.Zipkin/ZipkinExporter.cs	88.42% <ø> (ø)
...entation/ExportClient/OtlpGrpcTraceExportClient.cs	35.71% <0.00%> (-42.86%)	⬇️
...xporter.OpenTelemetryProtocol/OtlpTraceExporter.cs	36.36% <0.00%> (-40.91%)	⬇️
...lemetry/Internal/SelfDiagnosticsConfigRefresher.cs	86.53% <0.00%> (-2.89%)	⬇️
...metryProtocol/Implementation/ActivityExtensions.cs	88.73% <0.00%> (-2.71%)	⬇️
...nTelemetry/Internal/OpenTelemetrySdkEventSource.cs	74.13% <0.00%> (-2.59%)	⬇️
...heus/Implementation/PrometheusCollectionManager.cs	79.74% <0.00%> (-2.54%)	⬇️
...Telemetry/Internal/SelfDiagnosticsEventListener.cs	96.87% <0.00%> (-0.79%)	⬇️
...ter.ZPages/Implementation/ZPagesActivityTracker.cs	100.00% <0.00%> (+2.85%)	⬆️
...ZPages/Implementation/ZPagesExporterEventSource.cs	62.50% <0.00%> (+6.25%)	⬆️

[CodeBlanch]

@cijothomas @alanwest @reyang @utpilla

Was just looking at this while investigating something else. It didn't occur to me at the time, but we kind of introduced an issue doing this.

Previously we had a net5.0 target and code like this:

#if NET5_0_OR_GREATER
                using var response = this.httpClient.Send(request, CancellationToken.None);
#else

That became:

#if NET6_0_OR_GREATER
                using var response = this.httpClient.Send(request, CancellationToken.None);
#else

Users on .NET 5 upgrading to latest will still compile, they will just bind to the netstandard2.1 target. However, they have lost the optimization. So for them, upgrading was a perf hit/carried unforeseen consequences.

[alanwest]

Good find. Lesson is that in the future we need to align any drop of a target framework with the actual EOL date of the framework and consider which approach we're going to adopt here #3448