llama.cpp-cuda: auto updated to b5440-1

It has been built because:
* nvchecker detects the following updates: aur(0): b5439-1 -> b5440-1

Author: lilacbot

archlinuxcn/llama.cpp-cuda/PKGBUILD

@@ -2,7 +2,7 @@
 
 pkgname=llama.cpp-cuda
 _pkgname="${pkgname%-cuda}"
-pkgver=b5439
+pkgver=b5440
 pkgrel=1
 pkgdesc="Port of Facebook's LLaMA model in C/C++ (with NVIDIA CUDA optimizations)"
 arch=(x86_64 armv7h aarch64)
@@ -34,7 +34,7 @@ source=(
   llama.cpp.conf
   llama.cpp.service
 )
-sha256sums=('ffc627168d6a31e431f08b1939e28cacbc4582752157be5ca4294d15548c252f'
+sha256sums=('5765f0dbde44d4395834ec311b4fb1943b06f0efa9b536ea9c8cb21ac039cd7c'
             'SKIP'
             '53fa70cfe40cb8a3ca432590e4f76561df0f129a31b121c9b4b34af0da7c4d87'
             '0377d08a07bda056785981d3352ccd2dbc0387c4836f91fb73e6b790d836620d')