scripts : add build-zen.{sh,bat} helpers for AVX-512-capable CPUs (#1734)

Adds two thin one-liner-ish helpers that invoke `cmake` with the flag
set documented in docs/build.md "CPU build flags for AVX-512", so that
users on AMD Zen4 / Intel Sapphire Rapids+ hardware get the IQK
HAVE_FANCY_SIMD path activated without having to remember the five
relevant `GGML_AVX512_*=ON` options.

  scripts/build-zen.sh   - Linux / macOS bash wrapper
  scripts/build-zen.bat  - Windows MSVC wrapper (run from a
                           "x64 Native Tools Command Prompt")

Both default to a "build" output directory, both pass through to the
same cmake invocation, and both work alongside the existing build
options (no behavioural change to vanilla CMake builds).

.gitignore: added `!scripts/build-*.sh` / `!scripts/build-*.bat`
exceptions, in line with the existing `!build-info.sh` / `!build.zig`
exceptions, so the scripts directory build helpers don't get caught
by the broad `build*` artifact pattern.

This is a follow-up to #1729 — the docs section explains why these
flags matter, this PR makes them one command away.

.gitignore

@@ -49,6 +49,8 @@ build*
 !build-info.sh
 !build.zig
 !docs/build.md
+!scripts/build-*.sh
+!scripts/build-*.bat
 /libllama.so
 /llama-*
 /vulkan-shaders-gen

scripts/build-zen.bat

@@ -0,0 +1,27 @@
+@echo off
+REM CPU-only build helper for AVX-512-capable CPUs (AMD Zen4 / Intel
+REM Sapphire Rapids+) on Windows + MSVC. Enables the IQK GEMM kernels
+REM gated by HAVE_FANCY_SIMD (see docs\build.md "CPU build flags for AVX-512").
+REM
+REM Run from a Visual Studio "x64 Native Tools Command Prompt" so that
+REM cl.exe and the rest of the MSVC toolchain are on PATH.
+REM
+REM Usage:
+REM   scripts\build-zen.bat [build-dir]
+REM
+REM Default build directory is "build".
+
+setlocal
+
+if "%~1"=="" (set BUILD_DIR=build) else (set BUILD_DIR=%~1)
+
+cmake -B "%BUILD_DIR%" -G "NMake Makefiles" ^
+    -DCMAKE_BUILD_TYPE=Release ^
+    -DGGML_NATIVE=ON ^
+    -DGGML_AVX512=ON ^
+    -DGGML_AVX512_VBMI=ON ^
+    -DGGML_AVX512_VNNI=ON ^
+    -DGGML_AVX512_BF16=ON
+if errorlevel 1 exit /b 1
+
+cmake --build "%BUILD_DIR%" --config Release

scripts/build-zen.sh

@@ -0,0 +1,29 @@
+#!/bin/sh
+# CPU-only build helper for AVX-512-capable CPUs (AMD Zen4 / Intel
+# Sapphire Rapids+). Enables the IQK GEMM kernels gated by HAVE_FANCY_SIMD
+# (see docs/build.md "CPU build flags for AVX-512").
+#
+# Usage:
+#   ./scripts/build-zen.sh [build-dir]
+#
+# Default build directory is "build". A subsequent
+#
+#   objdump -d <build-dir>/bin/llama-cli | grep -c vpdpbusd
+#
+# should report a non-trivial count if VNNI was compiled in. The runtime
+# banner of any built binary will print "HAVE_FANCY_SIMD is defined" when
+# the AVX-512 quantized matmul path is active.
+
+set -e
+
+BUILD_DIR=${1:-build}
+
+cmake -B "$BUILD_DIR" \
+    -DCMAKE_BUILD_TYPE=Release \
+    -DGGML_NATIVE=ON \
+    -DGGML_AVX512=ON \
+    -DGGML_AVX512_VBMI=ON \
+    -DGGML_AVX512_VNNI=ON \
+    -DGGML_AVX512_BF16=ON
+
+cmake --build "$BUILD_DIR" --config Release -j