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: swolchok

examples/models/llama/runner/CMakeLists.txt

@@ -38,7 +38,7 @@ include(${EXECUTORCH_SRCS_FILE})
 list(TRANSFORM _llama_runner__srcs PREPEND "${EXECUTORCH_ROOT}/")
 
 target_include_directories(
-  extension_module INTERFACE ${_common_include_directories} ${EXECUTORCH_INCLUDE_DIRS}
+  extension_module INTERFACE ${_common_include_directories}
 )
 
 list(
@@ -82,6 +82,6 @@ set(llama_runner_deps executorch extension_data_loader extension_module
 target_link_libraries(llama_runner PUBLIC ${llama_runner_deps})
 
 target_include_directories(
-  llama_runner INTERFACE ${_common_include_directories} ${EXECUTORCH_INCLUDE_DIRS}
+  llama_runner INTERFACE ${_common_include_directories} ${EXECUTORCH_ROOT}
 )
 target_compile_options(llama_runner PUBLIC ${_preprocessor_flag})