Retry peer-reset timeouts in socket disposal tests on BSD/Apple

[MihaZupan]

Fixes #131990.

TcpReceiveSendGetsCanceledByDispose can time out waiting for the peer
to observe a reset after local cancellation has succeeded. This is
consistent with the documented BSD TCP reset-validation issue:
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=296594

The existing retry loop handles assertion failures, but not
TimeoutException. Convert only peer-receive timeouts on Apple/FreeBSD
send paths into logged assertion failures, allowing the existing
eight-attempt loop to retry with fresh sockets.

This preserves the peer-reset assertion. Timeouts waiting for local
cancellation or disposal still fail immediately. Also ensure the sending
socket is disposed on exceptional exits.

Validation

• Linux functional suite: 2,418 passed, 24 skipped, zero failures.
• Linux inline-completion test: passed.
• Helix macOS 27 ARM64: 174 iterations, 18,792 cases passed, zero skips.
  Iteration 174 encountered a peer-reset timeout in SendReceive_Apm;
  the same test then passed after retry, exercising the recovery path.

Helix results and captured retry

Note

This change and description were prepared with GitHub Copilot.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[dotnet-policy-service]

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

[Copilot]

🟢 Approval recommended

The change is test-only, narrowly scoped, and the retry behavior is constrained to the flaky peer-reset observation case on specific platforms.

Pull request overview

This PR makes the TcpReceiveSendGetsCanceledByDispose functional test more resilient on Apple/FreeBSD by allowing an existing retry loop to also retry peer-side reset observation timeouts (while keeping local cancellation/disposal timeouts as immediate failures), and ensures socket1 is disposed even on exceptional exits.

Changes:

• Wrap socket1 in a using to guarantee cleanup on all exits from the retry body.
• On Apple/FreeBSD send path, convert peer-receive TimeoutException into an assertion failure so RetryHelper can retry with fresh sockets.

File summaries

File	Description
src/libraries/System.Net.Sockets/tests/FunctionalTests/SendReceive/SendReceive.cs	Updates the retry/cleanup behavior in TcpReceiveSendGetsCanceledByDispose to handle peer-reset timeouts on Apple/FreeBSD and ensure socket1 disposal.

Review details

• Files reviewed: 1/1 changed files
• Comments generated: 1
• Review effort level: Lite

[Copilot]

🟢 Approval recommended

The change is narrowly scoped to test resilience with platform guards, preserves the core assertion, and improves resource cleanup without altering product code.

Review details

• Files reviewed: 1/1 changed files
• Comments generated: 0 new
• Review effort level: Lite

[MihaZupan]

/ba-g Quic failure is unrelated