Add interceptor to aggregate CCFB reports

[mengelbart]

Alternative but cleaner implementation for congestion control feedback receiver (#300). The main difference between this and #300 is that #300 organizes the feedback per stream, while this version organizes it as a single linear stream, which is easier to process. If one still needs to know which stream a packet was sent on, that information is still included in the reports for each packet.

[codecov]

Codecov Report

❌ Patch coverage is 78.27476% with 68 lines in your changes missing coverage. Please review.
✅ Project coverage is 79.78%. Comparing base (87546e8) to head (89b5044).
⚠️ Report is 5 commits behind head on master.

Files with missing lines	Patch %	Lines
pkg/rtpfb/interceptor.go	73.94%	25 Missing and 6 partials ⚠️
pkg/rtpfb/twcc_receiver.go	69.11%	21 Missing ⚠️
pkg/rtpfb/history.go	81.39%	11 Missing and 5 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #374      +/-   ##
==========================================
- Coverage   80.03%   79.78%   -0.26%     
==========================================
  Files          84       88       +4     
  Lines        4268     4581     +313     
==========================================
+ Hits         3416     3655     +239     
- Misses        681      741      +60     
- Partials      171      185      +14     

Flag	Coverage Δ
go	79.78% <78.27%> (-0.26%)	⬇️

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

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[JoTurk]

Just two fast comments / nits, I'll read the RFC again and will actually review :)

[JoTurk]

What will happen if latestArrival is never set? shouldn't we return 0, nil?

[mengelbart]

Good point, I think we should initialize latestArrival as latestArrival = referenceTime. That way it will be 0 if it is there were no packets received after the reference time.

[mengelbart]

Sorry, this doesn't work because we are actually looking for the latest arrival before the referenceTime. So you're right, we have to add a check and return 0.

[JoTurk]

I'm so sorry for taking long with the second review, The code is great, and I think it works as expected would love if we can add more tests, esp around concurrency & life-cycle, but ofc this can be done in the future.

Thank you. sorry again.

[mengelbart]

Thank you @JoTurk !