Custom OID by environment variable offset misalignment

[iMrYang]

Describe the bug

If environment variable is defined, the function oqs_patch_oids() modifies array oqs_oid_alg_list OID.
There is an incorrect offset value after the OQS_OID_RSA3072_MLDSA44 environment variable.
The next environment variable after OQS_OID_RSA3072_MLDSA44 is OQS_OID_MLDSA65, but in array oqs_oid_alg_list, after rsa3072_mldsa44 is mldsa44_pss2048, not mldsa65

const char *oqs_oid_alg_list[OQS_OID_CNT] = {
    ...
    
    "1.3.6.1.4.1.2.267.12.4.4",
    "mldsa44",
    "1.3.9999.7.1",
    "p256_mldsa44",
    "1.3.9999.7.2",
    "rsa3072_mldsa44",                  /* rsa3072_mldsa44,  */
    "2.16.840.1.114027.80.8.1.1",
    "mldsa44_pss2048",
    "2.16.840.1.114027.80.8.1.2",
    "mldsa44_rsa2048",
    "2.16.840.1.114027.80.8.1.3",
    "mldsa44_ed25519",
    "2.16.840.1.114027.80.8.1.4",
    "mldsa44_p256",
    "2.16.840.1.114027.80.8.1.5",
    "mldsa44_bp256",
    "1.3.6.1.4.1.2.267.12.6.5",
    "mldsa65",                          /* mldsa65  */
    "1.3.9999.7.3",
    "p384_mldsa65",

    ... 
};

int oqs_patch_oids(void)
{
    ...

    if ((envval = getenv("OQS_OID_MLDSA44")))
        oqs_oid_alg_list[14 + OQS_KEMOID_CNT] = envval;
    if ((envval = getenv("OQS_OID_P256_MLDSA44")))
        oqs_oid_alg_list[16 + OQS_KEMOID_CNT] = envval;
    if ((envval = getenv("OQS_OID_RSA3072_MLDSA44")))       /* rsa3072_mldsa44 */
        oqs_oid_alg_list[18 + OQS_KEMOID_CNT] = envval;
                                                            /* <-- here lose */
    if ((envval = getenv("OQS_OID_MLDSA65")))               /* mldsa65, offset is incorrect after this  */
        oqs_oid_alg_list[20 + OQS_KEMOID_CNT] = envval;
    if ((envval = getenv("OQS_OID_P384_MLDSA65")))
        oqs_oid_alg_list[22 + OQS_KEMOID_CNT] = envval;

    ...
}

[baentsch]

Thanks for this bug report, @iMrYang . This is indeed wrong.

The error is apparently caused by a missing code generation snippet for operating on composite algorithms that is present for the baseline OID list. Tagging @feventura fyi: You'd arguably be best placed to rectify this: Please let us know whether you'd have time to provide a PR for this or whether someone else should look into this.

[baentsch]

@feventura Please disregard my question above: It was easier to fix than I thought: Please check #419. @iMrYang please let me know whether you'd agree this solves the problem.

[iMrYang]

@baentsch I agree, at least the file oqsprov.c is as expected