Use atomics for write-dup mailbox state

[be0hhh]

Description

Use wolfSSL's portable atomic integer operations for the simple TLS write-dup error and KeyUpdate mailbox state when atomic operations are available.

The mutex remains responsible for lifetime and reference-count state. Configurations with DTLS 1.3 or post-handshake authentication retain the existing mutex mailbox path because they carry additional compound state and ownership.

This removes the explicit per-write mailbox mutex on the supported atomic path without changing the public API or the mailbox's level-triggered semantics.

Testing

• GCC RelWithDebInfo: write-dup API tests passed.
• Clang RelWithDebInfo: write-dup API tests passed.
• GCC with atomic operations disabled: existing write-dup tests passed and the atomic-only test was skipped.
• ASan and UBSan: write-dup API tests passed.
• DTLS 1.2 and DTLS 1.3 configuration: write-dup API tests passed on the retained mutex path.
• Post-handshake authentication configuration: library and unit-test binary built on the retained mutex path; the existing certificate-chain assertions fail identically on unmodified master.
• Full GCC unit suite: passed.
• Added a concurrent producer/consumer regression test for error publication and KeyUpdate take/post behavior.
• Ran the source-text hygiene check on all changed files.

ThreadSanitizer builds successfully after demoting its existing atomic_thread_fence warning, but its runtime exits before test execution in this WSL environment with unexpected memory mapping.

Checklist

• added tests
• updated/added doxygen: not applicable; internal-only change
• updated appropriate READMEs: not applicable
• Updated manual and documentation: not applicable

[wolfSSL-Bot]

Can one of the admins verify this patch?

[dgarske]

Hi @be0hhh , thank you for the code contribution. Can you tell us more about your project and interest in our library? To be an external contributor we require a signed agreement. If you are willing to do this please email support at wolfssl dot com and include your physical location and details about your project/interests and a link the your active PR(s).
Thanks, David Garske, wolfSSL

[github-actions]

MemBrowse Memory Report

No memory changes detected for:

• gcc-arm-cortex-m0plus
• gcc-arm-cortex-m3
• gcc-arm-cortex-m4
• gcc-arm-cortex-m4-baremetal
• gcc-arm-cortex-m4-crypto-only
• gcc-arm-cortex-m4-dtls13
• gcc-arm-cortex-m4-min-ecc
• gcc-arm-cortex-m4-openssl-compat
• gcc-arm-cortex-m4-pkcs7
• gcc-arm-cortex-m4-pq
• gcc-arm-cortex-m4-rsa-only
• gcc-arm-cortex-m4-sp-math
• gcc-arm-cortex-m4-tls12
• gcc-arm-cortex-m4-tls13
• gcc-arm-cortex-m7
• gcc-arm-cortex-m7-pq
• gcc-arm-cortex-m7-tls13
• linuxkm-pie
• linuxkm-standard
• stm32-sim-stm32h753

[be0hhh]

Thank you, David.
I am willing to review and sign the individual contributor agreement. I have emailed support@wolfssl.com.
Please let me know if you need any additional information.

Best regards,
be0hhh