ddtrace/tracer: add NewStartSpanConfig & WithStartSpanConfig API

[darccio]

What does this PR do?

Adds NewStartSpanConfig and WithStartSpanConfig functions to allow to use config structs.

Motivation

This API reduces the number of calls (in functional option form) in hot paths by giving the freedom to prepare a common span configuration in hot paths.

Microbenchmarks show that there are significant improvements:

Just running the functional options

❯ benchstat -col /scenario /tmp/with_config.txt
goos: darwin
goarch: arm64
pkg: github.com/DataDog/dd-trace-go/v2/ddtrace/tracer
          │     none     │         WithStartSpanConfig         │
          │    sec/op    │   sec/op     vs base                │
Config-10   169.65n ± 1%   70.17n ± 5%  -58.64% (p=0.000 n=10)

          │    none    │        WithStartSpanConfig         │
          │    B/op    │    B/op     vs base                │
Config-10   448.0 ± 0%   160.0 ± 0%  -64.29% (p=0.000 n=10)

          │    none    │        WithStartSpanConfig         │
          │ allocs/op  │ allocs/op   vs base                │
Config-10   5.000 ± 0%   2.000 ± 0%  -60.00% (p=0.000 n=10)

Creating spans (more realistic)

❯ benchstat -col /scenario /tmp/start_span.txt
goos: darwin
goarch: arm64
pkg: github.com/DataDog/dd-trace-go/v2/ddtrace/tracer
                   │    none     │        WithStartSpanConfig         │
                   │   sec/op    │   sec/op     vs base               │
StartSpanConfig-10   1.387µ ± 1%   1.327µ ± 2%  -4.36% (p=0.000 n=10)

                   │     none     │         WithStartSpanConfig          │
                   │     B/op     │     B/op      vs base                │
StartSpanConfig-10   2.390Ki ± 0%   2.108Ki ± 0%  -11.77% (p=0.000 n=10)

                   │    none    │        WithStartSpanConfig         │
                   │ allocs/op  │ allocs/op   vs base                │
StartSpanConfig-10   25.00 ± 0%   22.00 ± 0%  -12.00% (p=0.000 n=10)

Reviewer's Checklist

• Changed code has unit tests for its functionality at or near 100% coverage.
• 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.

For Datadog employees:

• If this PR touches code that handles credentials of any kind, such as Datadog API keys, I've requested a review from @DataDog/security-design-and-guidance.
• This PR doesn't touch any of that.

Unsure? Have a question? Request a review!

[pr-commenter]

Benchmarks

Benchmark execution time: 2023-11-20 14:12:31

Comparing candidate commit bbbe437 in PR branch dario.castane/AIT-8738/new-span-config with baseline commit 54feb34 in branch v2-dev.

Found 0 performance improvements and 0 performance regressions! Performance is the same for 39 metrics, 2 unstable metrics.

[ahmed-mez]

Looks good! Just added a few non-blocking comments.