Add HMAC-BLAKE2b and HMAC-BLAKE2s API functions

[holtrop-wolfssl]

Description

Add HMAC-BLAKE2b and HMAC-BLAKE2s API functions

Testing

Added unit tests with two KAT vectors for each HMAC-BLAKE2b and HMAC-BLAKE2s. Verified KAT vectors with both OpenSSL and Rust blake2 crate.

Checklist

• added tests
• updated/added doxygen
• updated appropriate READMEs
• Updated manual and documentation

[holtrop-wolfssl]

retest this please (org.jenkinsci.plugins.workflow.support.steps.AgentOfflineException: Unable to create live FilePath for wolf-linux-cloud-node-02zun3; wolf-linux-cloud-node-02zun3 was marked offline: Connection was broken)

[devin-ai-integration]

🛟 Devin Lifeguard found 2 likely issues in this PR

• use-forcezero snippet snippet: Replace both XMEMSET(x_key, 0, BLAKE2B_BLOCKBYTES); / XMEMSET(i_hash, 0, BLAKE2B_OUTBYTES); (and the corresponding BLAKE2s lines) with ForceZero(x_key, BLAKE2B_BLOCKBYTES); and ForceZero(i_hash, BLAKE2B_OUTBYTES);.
• limit-stack-usage snippet snippet: Move x_key and i_hash (and optionally the state) to heap-allocated buffers using the WOLFSSL_SMALL_STACK pattern (e.g., XMALLOC / XFREE) to keep per-call stack usage under 100 bytes.

@holtrop-wolfssl
please take a look at the above issues which Devin flagged. Devin will not fix these issues automatically.

[holtrop-wolfssl]

🛟 Devin Lifeguard found 2 likely issues in this PR

• use-forcezero snippet snippet: Replace both XMEMSET(x_key, 0, BLAKE2B_BLOCKBYTES); / XMEMSET(i_hash, 0, BLAKE2B_OUTBYTES); (and the corresponding BLAKE2s lines) with ForceZero(x_key, BLAKE2B_BLOCKBYTES); and ForceZero(i_hash, BLAKE2B_OUTBYTES);.
• limit-stack-usage snippet snippet: Move x_key and i_hash (and optionally the state) to heap-allocated buffers using the WOLFSSL_SMALL_STACK pattern (e.g., XMALLOC / XFREE) to keep per-call stack usage under 100 bytes.

@holtrop-wolfssl please take a look at the above issues which Devin flagged. Devin will not fix these issues automatically.

@douzzer / @SparkiDev - I had mostly copied and adapted @douzzer's kernel implementation for this. But would it make sense to you two to fix either/both of these Devin suggestions?