feat(botocore): add support for tagging bedrock message input/outputs

[Yun-Kim]

This PR adds support to the botocore integration's bedrock service to correctly tag input/output messages from Anthropic calls.

Previously bedrock's models only supported raw text prompts and returned text outputs. However, Anthropic's newest claude 3 supports a chat message API, which means we need to support that as well.

This change also switches to using tracer.trace() instead of tracer.start_span(..., activate=False) for bedrock spans, because the latter meant that bedrock spans would always be root spans (messing up parenting for traces containing non-root bedrock spans).
Additionally by getting rid of the activate=False argument, this means that bedrock spans will now continue to be the active span until the stream/body is completely consumed. Previously we allowed bedrock spans to not be active, but if other downstream operations happen in the bedrock span then they would not correctly be child spans of the bedrock spans.

Checklist

• Change(s) are motivated and described in the PR description
• Testing strategy is described if automated tests are not included in the PR
• Risks are described (performance impact, potential for breakage, maintainability)
• Change is maintainable (easy to change, telemetry, documentation)
• Library release note guidelines are followed or label changelog/no-changelog is set
• Documentation is included (in-code, generated user docs, public corp docs)
• Backport labels are set (if applicable)
• If this PR changes the public interface, I've notified @DataDog/apm-tees.
• If change touches code that signs or publishes builds or packages, or handles credentials of any kind, I've requested a review from @DataDog/security-design-and-guidance.

Reviewer Checklist

• Title is accurate
• All changes are related to the pull request's stated goal
• Description motivates each change
• Avoids breaking API changes
• Testing strategy adequately addresses listed risks
• Change is maintainable (easy to change, telemetry, documentation)
• Release note makes sense to a user of the library
• Author has acknowledged and discussed the performance implications of this PR as reported in the benchmarks PR comment
• Backport labels are set in a manner that is consistent with the release branch maintenance policy

[datadog-dd-trace-py-rkomorn]

Datadog Report

Branch report: yunkim/bedrock-anthropic-messages-support
Commit report: 682aded
Test service: dd-trace-py

✅ 0 Failed, 2990 Passed, 299 Skipped, 1h 31m 8.67s Total duration (4m 26.02s time saved)

[pr-commenter]

Benchmarks

Benchmark execution time: 2024-03-28 00:48:04

Comparing candidate commit 5c08ffd in PR branch yunkim/bedrock-anthropic-messages-support with baseline commit be880c1 in branch main.

Found 6 performance improvements and 0 performance regressions! Performance is the same for 195 metrics, 9 unstable metrics.

scenario:httppropagationextract-large_header_no_matches

• 🟩 max_rss_usage [-770.113KB; -549.209KB] or [-3.521%; -2.511%]

scenario:httppropagationextract-medium_header_no_matches

• 🟩 max_rss_usage [-757.711KB; -513.278KB] or [-3.464%; -2.346%]

scenario:httppropagationextract-wsgi_invalid_span_id_header

• 🟩 max_rss_usage [-812.180KB; -726.688KB] or [-3.712%; -3.322%]

scenario:httppropagationextract-wsgi_invalid_trace_id_header

• 🟩 max_rss_usage [-766.730KB; -673.833KB] or [-3.510%; -3.085%]

scenario:httppropagationextract-wsgi_large_header_no_matches

• 🟩 max_rss_usage [-765.449KB; -685.764KB] or [-3.497%; -3.133%]

scenario:httppropagationextract-wsgi_medium_header_no_matches

• 🟩 max_rss_usage [-789.064KB; -707.615KB] or [-3.609%; -3.237%]

[lievan]

Looks good, just have a question about the placeholder text for images