Update on "Reuse GELU implementation from PyTorch core"

kernels/optimized doesn't need to support embedded systems, so it can just take a header-only dep on PyTorch.

Note that, because we will pick up Sleef internally and ignore it
externally thanks to ATen vec, this PR gets to enable optimized GELU in OSS.

Testing: CI to make sure this doesn't break mobile build modes; happy to take advice on anything not currently covered that might break.

Differential Revision: [D66335522](https://our.internmc.facebook.com/intern/diff/D66335522/)

[ghstack-poisoned]

Author: Github Executorch

.ci/scripts/build_llama_android.sh

@@ -14,6 +14,8 @@ if [[ -z "${PYTHON_EXECUTABLE:-}" ]]; then
   PYTHON_EXECUTABLE=python3
 fi
 which "${PYTHON_EXECUTABLE}"
+SITE_PACKAGES="$(${PYTHON_EXECUTABLE} -c 'from distutils.sysconfig import get_python_lib; print(get_python_lib())')"
+CMAKE_PREFIX_PATH="${SITE_PACKAGES}/torch"
 
 install_executorch_and_backend_lib() {
   echo "Installing executorch and xnnpack backend"
@@ -27,6 +29,7 @@ install_executorch_and_backend_lib() {
     -DANDROID_ABI="${ANDROID_ABI}" \
     -DCMAKE_INSTALL_PREFIX=cmake-android-out \
     -DCMAKE_BUILD_TYPE=Release \
+    -DCMAKE_PREFIX_PATH="${CMAKE_PREFIX_PATH}" \
     -DEXECUTORCH_BUILD_EXTENSION_DATA_LOADER=ON \
     -DEXECUTORCH_BUILD_EXTENSION_MODULE=ON \
     -DEXECUTORCH_BUILD_EXTENSION_TENSOR=ON \
@@ -43,8 +46,6 @@ install_executorch_and_backend_lib() {
 build_llama_runner() {
     echo "Building llama runner for Android..."
     ANDROID_ABI=arm64-v8a
-    SITE_PACKAGES="$(${PYTHON_EXECUTABLE} -c 'from distutils.sysconfig import get_python_lib; print(get_python_lib())')"
-    CMAKE_PREFIX_PATH="${SITE_PACKAGES}/torch"
     cmake -DBUCK2="${BUCK2}" \
     -DCMAKE_TOOLCHAIN_FILE="$ANDROID_NDK"/build/cmake/android.toolchain.cmake  \
     -DANDROID_ABI="${ANDROID_ABI}" \

build/build_android_llm_demo.sh

@@ -7,6 +7,13 @@
 
 set -ex
 
+if [[ -z "${PYTHON_EXECUTABLE:-}" ]]; then
+  PYTHON_EXECUTABLE=python3
+fi
+which "${PYTHON_EXECUTABLE}"
+SITE_PACKAGES="$(${PYTHON_EXECUTABLE} -c 'from distutils.sysconfig import get_python_lib; print(get_python_lib())')"
+CMAKE_PREFIX_PATH="${SITE_PACKAGES}/torch"
+
 build_jar() {
   pushd extension/android
   ./gradlew build
@@ -36,6 +43,7 @@ build_android_native_library() {
   fi
 
   cmake . -DCMAKE_INSTALL_PREFIX="${CMAKE_OUT}" \
+    -DCMAKE_PREFIX_PATH="${CMAKE_PREFIX_PATH}" \
     -DCMAKE_TOOLCHAIN_FILE="${ANDROID_NDK}/build/cmake/android.toolchain.cmake" \
     -DANDROID_ABI="${ANDROID_ABI}" \
     -DANDROID_PLATFORM=android-26 \
@@ -69,6 +77,7 @@ build_android_native_library() {
     -DANDROID_ABI="${ANDROID_ABI}" \
     -DANDROID_PLATFORM=android-26 \
     -DCMAKE_INSTALL_PREFIX="${CMAKE_OUT}" \
+    -DCMAKE_PREFIX_PATH="${CMAKE_PREFIX_PATH}" \
     -DEXECUTORCH_ENABLE_LOGGING=ON \
     -DEXECUTORCH_LOG_LEVEL=Info \
     -DCMAKE_FIND_ROOT_PATH_MODE_PACKAGE=BOTH \