stop double-installing ExecuTorch in one-off linux jobs

setup-linux.sh already installs ExecuTorch with XNNPACK (and it passes use-pt-pinned-commit as it should).

Differential Revision: [D67996460](https://our.internmc.facebook.com/intern/diff/D67996460/)

[ghstack-poisoned]

Author: swolchok

.github/workflows/pull.yml

@@ -199,9 +199,6 @@ jobs:
 
         PYTHON_EXECUTABLE=python bash .ci/scripts/setup-linux.sh "cmake"
 
-        # install pybind
-        bash install_requirements.sh --pybind xnnpack
-
         # install Llava requirements
         bash examples/models/llama/install_requirements.sh
         bash examples/models/llava/install_requirements.sh
@@ -413,9 +410,6 @@ jobs:
 
         PYTHON_EXECUTABLE=python bash .ci/scripts/setup-linux.sh "cmake"
 
-        # install pybind
-        bash install_requirements.sh --pybind xnnpack
-
         # install phi-3-mini requirements
         bash examples/models/phi-3-mini/install_requirements.sh
 
@@ -440,9 +434,6 @@ jobs:
 
         PYTHON_EXECUTABLE=python bash .ci/scripts/setup-linux.sh "cmake"
 
-        # install pybind
-        bash install_requirements.sh --pybind xnnpack
-
         # install llama requirements
         bash examples/models/llama/install_requirements.sh
 
@@ -467,10 +458,6 @@ jobs:
 
         PYTHON_EXECUTABLE=python bash .ci/scripts/setup-linux.sh "cmake"
 
-        # install pybind
-        bash install_requirements.sh --pybind xnnpack
-
-        # install llama requirements
         bash examples/models/llama/install_requirements.sh
 
         # run eval_llama mmlu task
@@ -494,9 +481,6 @@ jobs:
 
         PYTHON_EXECUTABLE=python bash .ci/scripts/setup-linux.sh "cmake"
 
-        # install pybind
-        bash install_requirements.sh --pybind xnnpack
-
         # install llama requirements
         bash examples/models/llama/install_requirements.sh