[SmaCow666]

The RVV extrema kernels initialize their accumulators with FLT_MAX or -FLT_MAX, including in double precision. A finite input such as DZAMIN of one complex value {0x1p900, 0x1p899} therefore returns FLT_MAX instead of 0x1.8p900. The same initialization also clips all-positive-infinity minima and all-negative-infinity maxima.

Initialize both the reduction seed and vector accumulators from the first valid comparison value, after the existing empty/invalid-increment checks. This follows the scalar kernels and avoids introducing a finite sentinel into the reduction. Add twelve regression cases to the existing min/amin tests, including a finite double value immediately above FLT_MAX.

Local/QEMU validation used upstream 02f585444a0fba2505cfda77cc6684de008ea477, Clang 21.1.8, LP64D/32-bit BLAS integers, S/D/C/Z BLAS, pthreads with a two-thread limit, and QEMU 9.1.0:

• Complete builds from the pinned upstream tree for RISCV64_ZVL128B and RISCV64_ZVL256B, with each PR applied independently.
• Both targets: 153/153 unit tests and 1479/1479 extension tests pass, together with all enabled official Fortran BLAS and CBLAS Level-1/2/3 drivers in all four precisions.
• VLEN=512: ZVL128B passes 153/153 unit and 1479/1479 extension tests. ZVL256B has three unit failures and 34 extension failures, exactly matching the named failures in the unmodified library at the same VLEN and thread count; all twelve added regressions and the targeted API checks pass. This configuration is not an all-passing full-library result. Primary runs use the target's minimum VLEN and request two threads; VLEN512 runs request one.
• 43,200 boundary/classification checks across five RISC-V rounding modes, including 40,320 exception-flag comparisons, all passed. Arrays were read-only and bounded by inaccessible pages.
• Proposed tests linked to unmodified current archives fail 12 cases on ZVL128B and 9 on ZVL256B; both patched suites pass.
• Twenty Clang warning-as-error precision/NO_CBLAS compilations passed; NO_CBLAS test objects have no CBLAS references and their 22-test runners pass for both targets.
• GCC 11 scalar-kernel portability control: 22/22 tests passed (rv64gc).
• Library-member comparison confines changes to the affected kernels; exported symbol sets are unchanged. The patch passes git apply --check --whitespace=error-all against the pinned upstream tree.

BLAS was built with NO_LAPACK=1. A modern GCC RVV build and legacy C910V intrinsics have not been tested. No performance improvement is claimed.

Native RISC-V Linux validation (VLEN=256):

• The same independently patched binaries pass all four target/thread cells: RISCV64_ZVL128B and RISCV64_ZVL256B, each with one and two requested threads. Each cell passes 153/153 unit tests, 1479/1479 extension tests, and all 24 enabled official S/D/C/Z Fortran BLAS and CBLAS Level-1/2/3 drivers.
• This patch accounts for 120 native program executions; the shared unmodified-library baseline accounts for another 108. Both independently tested patches together account for 348 executions, rather than 348 per patch.
• 43,200 guarded boundary/classification calls and 40,320 exception-flag comparisons across five rounding modes pass. Each NO_CBLAS runner passes 22/22 tests. Proposed tests linked to unmodified libraries retain the expected 12 (ZVL128B) or 9 (ZVL256B) failures per thread setting.

Native and QEMU results are separate correctness evidence; no hardware performance measurement is claimed.