appsec: never start a new WAF Context when one is in context

[eliottness]

What does this PR do?

This PR does two things:

• Add a new FindOperation function in dyngo
• Use it to find an existing waf.ContextOperation in the current context so we don't start more than one at a time

Motivation

For example gorilla/mux and go-chi/chi are contribs that uses net/http under the hood and since orchestrion will be instrumenting both of them, we will have multiple WAF Contexts in the Operation start

Microbenchmarks

goos: linux
goarch: amd64
pkg: gopkg.in/DataDog/dd-trace-go.v1/internal/appsec/dyngo
cpu: 12th Gen Intel(R) Core(TM) i9-12900H
                                                  │   old.out    │               new.out                │
                                                  │    sec/op    │    sec/op     vs base                │
Events/emitting/stack=1/start_event-20              162.1n ± 19%   165.3n ± 19%        ~ (p=0.969 n=25)
Events/emitting/stack=1/start_+_finish_events-20    391.2n ±  9%   415.3n ± 12%   +6.16% (p=0.023 n=25)
Events/emitting/stack=2/start_event-20              165.0n ± 24%   182.4n ± 19%        ~ (p=0.488 n=25)
Events/emitting/stack=2/start_+_finish_events-20    413.0n ±  8%   455.1n ±  7%  +10.19% (p=0.009 n=25)
Events/emitting/stack=4/start_event-20              205.6n ± 20%   214.5n ± 22%        ~ (p=0.453 n=25)
Events/emitting/stack=4/start_+_finish_events-20    462.3n ±  8%   510.9n ±  6%  +10.51% (p=0.005 n=25)
Events/emitting/stack=8/start_event-20              260.2n ± 14%   276.9n ± 14%        ~ (p=0.255 n=25)
Events/emitting/stack=8/start_+_finish_events-20    594.3n ±  6%   664.6n ±  6%  +11.83% (p=0.000 n=25)
Events/emitting/stack=16/start_event-20             368.9n ± 10%   411.7n ± 11%  +11.60% (p=0.041 n=25)
Events/emitting/stack=16/start_+_finish_events-20   831.9n ±  4%   936.6n ±  5%  +12.59% (p=0.000 n=25)
Events/emitting/stack=32/start_event-20             611.7n ±  7%   672.7n ±  7%   +9.97% (p=0.000 n=25)
Events/emitting/stack=32/start_+_finish_events-20   1.309µ ±  3%   1.470µ ±  2%  +12.30% (p=0.000 n=25)
Events/emitting/stack=64/start_event-20             1.058µ ±  5%   1.212µ ±  3%  +14.56% (p=0.000 n=25)
Events/emitting/stack=64/start_+_finish_events-20   2.238µ ±  1%   2.539µ ±  2%  +13.45% (p=0.000 n=25)
Events/registering/start_event-20                   73.26n ± 80%   84.20n ± 54%        ~ (p=0.985 n=25)
Events/registering/finish_event-20                  81.32n ± 29%   83.94n ± 29%        ~ (p=0.497 n=25)
geomean                                             383.6n         420.4n         +9.59%

                                                  │   old.out    │                new.out                 │
                                                  │     B/op     │     B/op      vs base                  │
Events/emitting/stack=1/start_event-20              128.0 ±   0%   128.0 ±   0%        ~ (p=1.000 n=25) ¹
Events/emitting/stack=1/start_+_finish_events-20    128.0 ±   0%   144.0 ±   0%  +12.50% (p=0.000 n=25)
Events/emitting/stack=2/start_event-20              128.0 ±   0%   128.0 ±   0%        ~ (p=1.000 n=25) ¹
Events/emitting/stack=2/start_+_finish_events-20    128.0 ±   0%   144.0 ±   0%  +12.50% (p=0.000 n=25)
Events/emitting/stack=4/start_event-20              128.0 ±   0%   128.0 ±   0%        ~ (p=1.000 n=25) ¹
Events/emitting/stack=4/start_+_finish_events-20    128.0 ±   0%   144.0 ±   0%  +12.50% (p=0.000 n=25)
Events/emitting/stack=8/start_event-20              128.0 ±   0%   128.0 ±   0%        ~ (p=1.000 n=25) ¹
Events/emitting/stack=8/start_+_finish_events-20    128.0 ±   0%   144.0 ±   0%  +12.50% (p=0.000 n=25)
Events/emitting/stack=16/start_event-20             128.0 ±   0%   128.0 ±   0%        ~ (p=1.000 n=25) ¹
Events/emitting/stack=16/start_+_finish_events-20   128.0 ±   0%   144.0 ±   0%  +12.50% (p=0.000 n=25)
Events/emitting/stack=32/start_event-20             128.0 ±   0%   128.0 ±   0%        ~ (p=1.000 n=25) ¹
Events/emitting/stack=32/start_+_finish_events-20   128.0 ±   0%   144.0 ±   0%  +12.50% (p=0.000 n=25)
Events/emitting/stack=64/start_event-20             128.0 ±   0%   128.0 ±   0%        ~ (p=1.000 n=25) ¹
Events/emitting/stack=64/start_+_finish_events-20   128.0 ±   0%   144.0 ±   0%  +12.50% (p=0.000 n=25)
Events/registering/start_event-20                   16.00 ± 662%   16.00 ± 712%        ~ (p=0.962 n=25)
Events/registering/finish_event-20                  16.00 ± 644%   16.00 ± 606%        ~ (p=0.954 n=25)
geomean                                             98.70          103.9          +5.29%
¹ all samples are equal

                                                  │  old.out   │               new.out                │
                                                  │ allocs/op  │ allocs/op   vs base                  │
Events/emitting/stack=1/start_event-20              2.000 ± 0%   2.000 ± 0%        ~ (p=1.000 n=25) ¹
Events/emitting/stack=1/start_+_finish_events-20    2.000 ± 0%   3.000 ± 0%  +50.00% (p=0.000 n=25)
Events/emitting/stack=2/start_event-20              2.000 ± 0%   2.000 ± 0%        ~ (p=1.000 n=25) ¹
Events/emitting/stack=2/start_+_finish_events-20    2.000 ± 0%   3.000 ± 0%  +50.00% (p=0.000 n=25)
Events/emitting/stack=4/start_event-20              2.000 ± 0%   2.000 ± 0%        ~ (p=1.000 n=25) ¹
Events/emitting/stack=4/start_+_finish_events-20    2.000 ± 0%   3.000 ± 0%  +50.00% (p=0.000 n=25)
Events/emitting/stack=8/start_event-20              2.000 ± 0%   2.000 ± 0%        ~ (p=1.000 n=25) ¹
Events/emitting/stack=8/start_+_finish_events-20    2.000 ± 0%   3.000 ± 0%  +50.00% (p=0.000 n=25)
Events/emitting/stack=16/start_event-20             2.000 ± 0%   2.000 ± 0%        ~ (p=1.000 n=25) ¹
Events/emitting/stack=16/start_+_finish_events-20   2.000 ± 0%   3.000 ± 0%  +50.00% (p=0.000 n=25)
Events/emitting/stack=32/start_event-20             2.000 ± 0%   2.000 ± 0%        ~ (p=1.000 n=25) ¹
Events/emitting/stack=32/start_+_finish_events-20   2.000 ± 0%   3.000 ± 0%  +50.00% (p=0.000 n=25)
Events/emitting/stack=64/start_event-20             2.000 ± 0%   2.000 ± 0%        ~ (p=1.000 n=25) ¹
Events/emitting/stack=64/start_+_finish_events-20   2.000 ± 0%   3.000 ± 0%  +50.00% (p=0.000 n=25)
Events/registering/start_event-20                   1.000 ± 0%   1.000 ± 0%        ~ (p=1.000 n=25) ¹
Events/registering/finish_event-20                  1.000 ± 0%   1.000 ± 0%        ~ (p=1.000 n=25) ¹
geomean        

Reviewer's Checklist

• Changed code has unit tests for its functionality at or near 100% coverage.
• System-Tests covering this feature have been added and enabled with the va.b.c-dev version tag.
• There is a benchmark for any new code, or changes to existing code.
• If this interacts with the agent in a new way, a system test has been added.
• Add an appropriate team label so this PR gets put in the right place for the release notes.
• Non-trivial go.mod changes, e.g. adding new modules, are reviewed by @DataDog/dd-trace-go-guild.

Unsure? Have a question? Request a review!

[pr-commenter]

Benchmarks

Benchmark execution time: 2024-11-26 10:33:13

Comparing candidate commit 4e23ec9 in PR branch eliottness/deduplicated-the-right-operations with baseline commit e9a64d3 in branch main.

Found 2 performance improvements and 0 performance regressions! Performance is the same for 57 metrics, 0 unstable metrics.

scenario:BenchmarkSetTagMetric-24

• 🟩 execution_time [-5.122ns; -3.118ns] or [-4.257%; -2.591%]

scenario:BenchmarkSetTagStringer-24

• 🟩 execution_time [-6.118ns; -3.322ns] or [-4.296%; -2.333%]