chore: Ensure consistent DT header insertion/replacement logic across all wrappers

[tippmar-nr]

Description

Validates and updates all distributed trace header insertion/replacement logic to ensure consistent behavior everywhere. Only SQS, RestSharp and the (very old) Wcf3 ChannelFactory wrapper were implemented inconsistently.

Adds unit and integration tests that ensure the replacement logic works correctly everywhere.

See the related Jira ticket and the Google doc referenced in it for a report on the entire inventory of DT header insertion across the .NET agent.

One change that's important to note is in the RestSharp AppendHeaders wrapper - header replacement was moved to the onComplete delegate to ensure anything added by the real AppendHeaders call would be correctly replaced if there was a collision. This was potentially a bug in the original implementation, but would have been very unlikely to occur.

Some wrapper implementations could not be tested easily or did not need a test because of the way the underlying library handles headers:

• NServiceBus — uses dictionary indexer headers[key] = value which inherently replaces. Headers are managed internally
  by the NServiceBus pipeline, so pre-populating DT headers before the agent's hook would require a custom pipeline
  behavior.
• MassTransit (current and legacy) — uses SendHeaders.Set(key, value) which is MassTransit's own API that replaces by
  design. Pre-populating would require a custom ISendObserver or IFilter.
• WCF NetTcp — the Message.Headers are not directly accessible before the agent's BeforeSendRequest inspector runs in
  the WCF client programming model, so there's no natural way for pre-existing DT headers to appear. The fix (FindHeader + RemoveAt + Add) was validated by code review.

Author Checklist

• Unit tests, Integration tests, and Unbounded tests completed
• Performance testing completed with satisfactory results (if required)

Reviewer Checklist

• Perform code review
• Pull request was adequately tested (new/existing tests, performance tests)

[nr-ahemsath]

Overall this looks great, I just have a minor concern about magic string values being hardcoded in both test code and exerciser code; this may be a fundamental limitation of how our tests are structured, but it least should be documented very clearly that those magic values have to match exactly in order for the tests to actually be verifying what we want them to.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 81.81%. Comparing base (24819c4) to head (5f5fd3b).
⚠️ Report is 4 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3512      +/-   ##
==========================================
+ Coverage   81.79%   81.81%   +0.01%     
==========================================
  Files         508      508              
  Lines       34220    34228       +8     
  Branches     4040     4040              
==========================================
+ Hits        27990    28003      +13     
+ Misses       5265     5257       -8     
- Partials      965      968       +3     

Flag	Coverage Δ
Agent	82.80% <100.00%> (+0.02%)	⬆️
Profiler	71.75% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
...ions/NewRelic.Agent.Extensions/AwsSdk/SqsHelper.cs	91.42% <100.00%> (+1.10%)	⬆️

... and 3 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.