Add bulk of telemetry V2 implementation

[andrewlock]

Summary of changes

Adds the bulk of the telemetry V2 implementation

• Adds v2 versions of the telemetry DTOs (where they have changed from v1)
• Adds v2 versions of collectors
• Doesn't enable actually using the v2 implementation yet.

Reason for change

This is the bulk of the actual v2 telemetry implementation.

Implementation details

I mean... so many details... Probably best to review commit-by-commit 😬

Test coverage

Unit tests for most things, integration tests will follow in a subsequent PR when we actually enable it.

Other details

Stacked on #4187
Stacked on #4180

[datadog-ddstaging]

Datadog Report

Branch report: andrew/telemetry/v2
Commit report: bfc6e0f

❄️ dd-trace-dotnet: 0 Failed, 1 New Flaky, 280928 Passed, 913 Skipped, 40m 33.07s Wall Time

New Flaky Tests (1)

• CheckSpanContextAreAttachedForCpuProfiler - Datadog.Profiler.IntegrationTests.CodeHotspot.CodeHotspotTest - Last Failure

  Expand for error

   Assert.NotEmpty() Failure
  

[pierotibou]

Thanks and thanks for all the comments, that was helpful

[andrewlock]

Benchmarks Report 🐌

Benchmarks for #4188 compared to master:

• All benchmarks have the same speed
• All benchmarks have the same allocations

The following thresholds were used for comparing the benchmark speeds:

• Mann–Whitney U test with statistical test for significance of 5%
• Only results indicating a difference greater than 10% and 0.3 ns are considered.

Allocation changes below 0.5% are ignored.

Benchmark details

[andrewlock]

Execution-Time Benchmarks Report ⏱️

Execution-time results for samples comparing the following branches/commits:

• This PR (4188)
• master

Execution-time benchmarks measure the whole time it takes to execute a program. And are intended to measure the one-off costs. Cases where the execution time results for the PR are worse than latest master results are shown in red. The following thresholds were used for comparing the execution times:

• Welch test with statistical test for significance of 5%
• Only results indicating a difference greater than 5% and 5 ms are considered.

Note that these results are based on a single point-in-time result for each branch. For full results, see the dashboard.

Graphs show the p99 interval based on the mean and StdDev of the test run, as well as the mean value of the run (shown as a diamond below the graph).

gantt
    title Execution time (ms) FakeDbCommand (.NET Framework 4.6.2) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4188) - mean (3,028ms)  : 2927, 3130
     .   : milestone, 3028,
    master - mean (3,025ms)  : 2925, 3125
     .   : milestone, 3025,

    section CallTarget+Inlining+NGEN
    This PR (4188) - mean (3,767ms)  : 3715, 3819
     .   : milestone, 3767,
    master - mean (3,743ms)  : 3672, 3815
     .   : milestone, 3743,

Loading

gantt
    title Execution time (ms) FakeDbCommand (.NET Core 3.1) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4188) - mean (3,154ms)  : 3049, 3260
     .   : milestone, 3154,
    master - mean (3,148ms)  : 2995, 3301
     .   : milestone, 3148,

    section CallTarget+Inlining+NGEN
    This PR (4188) - mean (3,596ms)  : 3486, 3707
     .   : milestone, 3596,
    master - mean (3,598ms)  : 3534, 3662
     .   : milestone, 3598,

Loading

gantt
    title Execution time (ms) FakeDbCommand (.NET 6) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4188) - mean (3,143ms)  : 3036, 3250
     .   : milestone, 3143,
    master - mean (3,131ms)  : 3016, 3246
     .   : milestone, 3131,

    section CallTarget+Inlining+NGEN
    This PR (4188) - mean (3,569ms)  : 3517, 3622
     .   : milestone, 3569,
    master - mean (3,583ms)  : 3523, 3643
     .   : milestone, 3583,

Loading

gantt
    title Execution time (ms) HttpMessageHandler (.NET Framework 4.6.2) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4188) - mean (189ms)  : 185, 192
     .   : milestone, 189,
    master - mean (188ms)  : 184, 191
     .   : milestone, 188,

    section CallTarget+Inlining+NGEN
    This PR (4188) - mean (1,021ms)  : 962, 1081
     .   : milestone, 1021,
    master - mean (1,023ms)  : 960, 1085
     .   : milestone, 1023,

Loading

gantt
    title Execution time (ms) HttpMessageHandler (.NET Core 3.1) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4188) - mean (369ms)  : 362, 376
     .   : milestone, 369,
    master - mean (367ms)  : 362, 371
     .   : milestone, 367,

    section CallTarget+Inlining+NGEN
    This PR (4188) - mean (1,133ms)  : 1100, 1165
     .   : milestone, 1133,
    master - mean (1,133ms)  : 1109, 1157
     .   : milestone, 1133,

Loading

gantt
    title Execution time (ms) HttpMessageHandler (.NET 6) 
    dateFormat  X
    axisFormat %s
    todayMarker off
    section Baseline
    This PR (4188) - mean (357ms)  : 351, 364
     .   : milestone, 357,
    master - mean (355ms)  : 351, 359
     .   : milestone, 355,

    section CallTarget+Inlining+NGEN
    This PR (4188) - mean (1,100ms)  : 1079, 1122
     .   : milestone, 1100,
    master - mean (1,100ms)  : 1061, 1138
     .   : milestone, 1100,

Loading

[andrewlock]

Throughput/Crank Report:zap:

Throughput results for AspNetCoreSimpleController comparing the following branches/commits:

• This PR (4188)
• master
• benchmarks/2.9.0
• benchmarks/2.31.0

Cases where throughput results for the PR are worse than latest master (5% drop or greater), results are shown in red.

Note that these results are based on a single point-in-time result for each branch. For full results, see one of the many, many dashboards!

gantt
    title Throughput Linux x64 (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (4188) (11.507M)   : 0, 11507459
    master (10.219M)   : 0, 10219339
    benchmarks/2.31.0 (11.338M)   : 0, 11338235
    benchmarks/2.9.0 (11.162M)   : 0, 11161538

    section Automatic
    This PR (4188) (8.066M)   : 0, 8066326
    master (7.156M)   : 0, 7155872
    benchmarks/2.31.0 (7.431M)   : 0, 7431225
    benchmarks/2.9.0 (8.099M)   : 0, 8099075

    section Trace stats
    master (7.149M)   : 0, 7148816
    benchmarks/2.31.0 (7.663M)   : 0, 7663176

    section Manual
    This PR (4188) (10.190M)   : 0, 10189635
    master (9.088M)   : 0, 9087972
    benchmarks/2.31.0 (7.351M)   : 0, 7350869

    section Manual + Automatic
    This PR (4188) (7.802M)   : 0, 7802267
    master (6.780M)   : 0, 6780456
    benchmarks/2.31.0 (5.741M)   : 0, 5741089

    section Version Conflict
    master (6.228M)   : 0, 6228238
    benchmarks/2.31.0 (6.913M)   : 0, 6913078

Loading

gantt
    title Throughput Linux arm64 (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (4188) (9.691M)   : 0, 9690581
    master (9.389M)   : 0, 9389432
    benchmarks/2.31.0 (9.679M)   : 0, 9678682
    benchmarks/2.9.0 (9.705M)   : 0, 9704781

    section Automatic
    This PR (4188) (6.719M)   : 0, 6718957
    master (6.921M)   : 0, 6920892
    benchmarks/2.31.0 (6.855M)   : 0, 6854574

    section Trace stats
    master (6.982M)   : 0, 6982075
    benchmarks/2.31.0 (6.924M)   : 0, 6924210

    section Manual
    This PR (4188) (8.540M)   : 0, 8539930
    master (8.479M)   : 0, 8479114
    benchmarks/2.31.0 (8.386M)   : 0, 8386139

    section Manual + Automatic
    This PR (4188) (6.531M)   : 0, 6530515
    master (6.645M)   : 0, 6644683
    benchmarks/2.31.0 (6.325M)   : 0, 6325089

    section Version Conflict
    master (6.046M)   : 0, 6045683
    benchmarks/2.31.0 (5.955M)   : 0, 5954718

Loading

gantt
    title Throughput Windows x64 (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (4188) (10.398M)   : 0, 10398370
    master (10.296M)   : 0, 10296222
    benchmarks/2.31.0 (10.320M)   : 0, 10319575
    benchmarks/2.9.0 (9.827M)   : 0, 9827121

    section Automatic
    This PR (4188) (7.505M)   : 0, 7505364
    master (7.512M)   : 0, 7511831
    benchmarks/2.31.0 (7.410M)   : 0, 7410117
    benchmarks/2.9.0 (7.246M)   : 0, 7246397

    section Trace stats
    master (7.455M)   : 0, 7454826
    benchmarks/2.31.0 (7.559M)   : 0, 7559264

    section Manual
    This PR (4188) (9.185M)   : 0, 9185112
    master (9.074M)   : 0, 9074178
    benchmarks/2.31.0 (9.387M)   : 0, 9386685

    section Manual + Automatic
    This PR (4188) (7.238M)   : 0, 7237739
    master (7.175M)   : 0, 7174810
    benchmarks/2.31.0 (7.176M)   : 0, 7175876

    section Version Conflict
    master (6.581M)   : 0, 6580964
    benchmarks/2.31.0 (6.446M)   : 0, 6446354

Loading

gantt
    title Throughput Linux x64 (ASM) (Total requests) 
    dateFormat  X
    axisFormat %s
    section Baseline
    This PR (4188) (7.374M)   : 0, 7374306
    master (7.303M)   : 0, 7302577
    benchmarks/2.31.0 (7.475M)   : 0, 7475399
    benchmarks/2.9.0 (7.748M)   : 0, 7748029

    section No attack
    This PR (4188) (2.335M)   : 0, 2335401
    master (2.387M)   : 0, 2386831
    benchmarks/2.31.0 (2.389M)   : 0, 2388686
    benchmarks/2.9.0 (3.274M)   : 0, 3273699

    section Attack
    This PR (4188) (2.016M)   : 0, 2015885
    master (1.993M)   : 0, 1992843
    benchmarks/2.31.0 (2.039M)   : 0, 2038865
    benchmarks/2.9.0 (2.598M)   : 0, 2597950

    section Blocking
    This PR (4188) (4.173M)   : 0, 4173011
    master (4.016M)   : 0, 4015526
    benchmarks/2.31.0 (4.079M)   : 0, 4079419

Loading

[bouwkast]

Looks good to me overall! I just had a few nitpick/curiosity questions

Tests looked good to me - some of the tests files were a bit large to follow (like tracer/test/Datadog.Trace.Tests/Telemetry/TelemetryControllerV2Tests.cs) but I don't have any suggestions and for the most part there was similar setup and the comments helped

[bouwkast]

Could 2000 be swapped for maxPerMessage here?

Suggested change

	? dependencies.Skip(skip).Take(2000).ToList()
	? dependencies.Skip(skip).Take(maxPerMessage).ToList()

[andrewlock]

thanks, will fix in follow-up to avoid the CI hit!