🔖 Bump llama.cpp-hip version to b6351-1

PKGBUILD

@@ -8,6 +8,7 @@ pkgrel=1
 pkgdesc="Port of Facebook's LLaMA model in C/C++ (with AMD ROCm optimizations)"
 arch=(x86_64 armv7h aarch64)
 url='https://github.com/ggml-org/llama.cpp'
+api_url='https://api.github.com/repos/ggml-org/llama.cpp/releases/latest'
 license=('MIT')
 depends=(
   curl
@@ -39,12 +40,18 @@ sha256sums=()
 
 prepare() {
   cd "$srcdir"
-  git clone --depth 1 --single-branch --branch master "${url}" "${_pkgname}"
+
+  local _latest_tag
+  _latest_tag=$(curl -s "${api_url}" | grep '"tag_name":' | sed -E 's/.*"([^"]+)".*/\1/')
+
+  # 仅使用刚获取的标签进行浅克隆, 避免脱离 Head 提示
+  msg2 "Cloning the latest release tag: ${_latest_tag}"
+  git -c advice.detachedHead=false clone --depth 1 --single-branch --branch "${_latest_tag}" "${url}" "${_pkgname}"
 }
 
 pkgver() {
   # 使用 API 获取最新发布的标签
-  curl -s "https://api.github.com/repos/ggml-org/llama.cpp/releases/latest" | \
+  curl -s "${api_url}" | \
     grep '"tag_name":' | \
     sed -E 's/.*"([^"]+)".*/\1/'
 }