[HTTP] Fix PING test for 0 lifetime connection

[ManickaP]

Fixes #116862

• Separates the test for 0 lifetime, it has to have a different logic since the connection will get closed after the first request.
• Fixes PING frame counting to exclude RTT related PINGs.

[dotnet-policy-service]

Tagging subscribers to this area: @dotnet/ncl
See info in area-owners.md if you want to be subscribed.

[ManickaP]

/azp run runtime-libraries-coreclr outerloop

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[Copilot]

Pull Request Overview

This PR improves HTTP/2 keep-alive ping tests by handling zero‐lifetime connections separately and excluding RTT-related pings from the counter.

• Adds a dedicated test (KeepAliveConfigured_KeepAlivePingsAreSentEvenWith0ConnectionLifetime) to verify pings are still sent when PooledConnectionLifetime is zero.
• Simplifies the existing policy-based test by removing the lifetime parameter cases.
• Modifies ProcessIncomingFramesAsync to only count real pings (pingFrame.Data >= 0), ignoring RTT pings.

Comments suppressed due to low confidence (1)

src/libraries/System.Net.Http/tests/FunctionalTests/SocketsHttpHandlerTest.Http2KeepAlivePing.cs:342

• Consider adding a unit test that simulates RTT-related PING frames (with Data < 0) to verify that these frames are correctly excluded from the ping counter.

                            if (pingFrame.Data >= 0)

[MihaZupan]

Thanks

[ManickaP]

/ba-g failures are unrelated