Add sha3 and icsf tests

[1000TurquoisePogs]

Proposed changes

Fix three bugs in the ICSF digest wrapper (icsf.h/icsf.c) and add SHA-3 algorithm support.

Bug fixes

1. SHA-256 typo (icsf.c): The icsfDigestInit rule array for ICSF_DIGEST_SHA256 contained "SHA246" instead of "SHA256". This caused SHA-256 hashing to fail at the ICSF callable service level with a bad rule array keyword. Fixed by correcting the string to "SHA256 FIRST ".

2. Hash buffer overflow (icsf.h): The hash field in ICSFDigest was declared as char hash[32], which is too small for SHA-384 (48 bytes) and SHA-512 (64 bytes). Writing the hash output would overflow into adjacent struct memory. Fixed by increasing to char hash[64].

3. Context buffer too small for SHA-3 (icsf.h): ICSF_HASH_CONTEXT_LENGTH was 128 bytes, which is insufficient for SHA-3's Keccak state (requires ~200+ bytes). Increased to 512 bytes to accommodate all current and foreseeable ICSF algorithm context sizes.

New feature: SHA-3 support

Added SHA-3 (Keccak) family algorithms to the ICSF digest wrapper:

Constant	Value	ICSF keyword	Hash length
ICSF_DIGEST_SHA3_224	7	SHA3-224	28 bytes
ICSF_DIGEST_SHA3_256	8	SHA3-256	32 bytes
ICSF_DIGEST_SHA3_384	9	SHA3-384	48 bytes
ICSF_DIGEST_SHA3_512	10	SHA3-512	64 bytes

Cases added to all three streaming functions: icsfDigestInit, icsfDigestUpdate, icsfDigestFinish.

Files changed

File	Change
h/icsf.h	Fixed ICSF_HASH_CONTEXT_LENGTH (128->512), hash[32]->hash[64], added SHA-3 constants
c/icsf.c	Fixed "SHA246" typo, added SHA-3 cases to Init/Update/Finish

Type of change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)

PR Checklist

• If the changes in this PR are meant for the next release / mainline, this PR targets the "staging" branch.
• My code follows the style guidelines of this project (see: Contributing guideline)
• I have commented my code, particularly in hard-to-understand areas
• Relevant update to CHANGELOG.md
• My changes generate no new warnings

Testing

Try out tests/icsftest.c

Further comments

The SHA-256 typo ("SHA246") has been present since the original code. Any existing callers using ICSF_DIGEST_SHA256 through this wrapper were silently getting ICSF errors. The SHA-384/SHA-512 buffer overflow was a latent bug that could cause memory corruption when those algorithms were used.

The context length increase from 128 to 512 bytes increases the ICSFDigest struct size. Callers allocating this struct on the stack should have sufficient stack space (the struct is now approximately 640 bytes total).

[sonarqubecloud]

Quality Gate failed

Failed conditions
1 Security Hotspot
6.8% Duplication on New Code (required ≤ 3%)

See analysis details on SonarQube Cloud

[Martin-Zeithaml]

Could you, please, add the PR or issue number?