Update on "[Executorch] Add quantized kv cache to oss ci"

Fixes to make sure quantized kv cache works in oss

Differential Revision: [D66269487](https://our.internmc.facebook.com/intern/diff/D66269487/)

[ghstack-poisoned]

Author: kimishpatel

examples/models/llama/source_transformation/quantized_kv_cache.py

@@ -22,6 +22,8 @@
 
     import executorch
 
+    from executorch.extension.pybindings import portable_lib  # noqa # usort: skip
+
     # Ideally package is installed in only one location but usage of
     # PYATHONPATH can result in multiple locations.
     # ATM this is mainly used in CI for qnn runner. Will need to revisit this