Resolve need to pin `protobuf` package to older version

[areusch]

Currently we are unable to rebuild docker images because protobuf released version 4.20.1 which i believe broke the older version of TensorFlow we're using. Likely the correct long-term fix here is to update our TensorFlow dependency to 2.9--I think some descriptors in TensorFlow were compiled against an older version of the package.

$ docker/build.sh ci_cpu
---> Running in 3bfc0bbd44eb
The command '/bin/sh -c bash /install/ubuntu_install_tflite.sh' returned a non-zero code: 1
ERROR: docker build failed.
$ docker commit 3bfc0bbd44eb test-build
$ docker run -it test-build bash
root@fae8461bbf76:/# python3
Python 3.7.5 (default, Dec  9 2021, 17:04:37)
[GCC 8.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/lib/python3.7/dist-packages/tensorflow/__init__.py", line 41, in <module>
    from tensorflow.python.tools import module_util as _module_util
  File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/__init__.py", line 40, in <module>
    from tensorflow.python.eager import context
  File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/eager/context.py", line 32, in <module>
    from tensorflow.core.framework import function_pb2
  File "/usr/local/lib/python3.7/dist-packages/tensorflow/core/framework/function_pb2.py", line 16, in <module>
    from tensorflow.core.framework import attr_value_pb2 as tensorflow_dot_core_dot_framework_dot_attr__value__pb2
  File "/usr/local/lib/python3.7/dist-packages/tensorflow/core/framework/attr_value_pb2.py", line 16, in <module>
    from tensorflow.core.framework import tensor_pb2 as tensorflow_dot_core_dot_framework_dot_tensor__pb2
  File "/usr/local/lib/python3.7/dist-packages/tensorflow/core/framework/tensor_pb2.py", line 16, in <module>
    from tensorflow.core.framework import resource_handle_pb2 as tensorflow_dot_core_dot_framework_dot_resource__handle__pb2
  File "/usr/local/lib/python3.7/dist-packages/tensorflow/core/framework/resource_handle_pb2.py", line 16, in <module>
    from tensorflow.core.framework import tensor_shape_pb2 as tensorflow_dot_core_dot_framework_dot_tensor__shape__pb2
  File "/usr/local/lib/python3.7/dist-packages/tensorflow/core/framework/tensor_shape_pb2.py", line 42, in <module>
    serialized_options=None, file=DESCRIPTOR),
  File "/usr/local/lib/python3.7/dist-packages/google/protobuf/descriptor.py", line 560, in __new__
    _message.Message._CheckCalledFromGeneratedFile()
TypeError: Descriptors cannot not be created directly.
If this call came from a _pb2.py file, your generated code is out of date and must be regenerated with protoc >= 3.19.0.
If you cannot immediately regenerate your protos, some other possible workarounds are:
 1. Downgrade the protobuf package to 3.20.x or lower.
 2. Set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python (but this will use pure-Python parsing and will be much slower).

More information: https://developers.google.com/protocol-buffers/docs/news/2022-05-06#python-updates

cc @driazati @leandron @tqchen @junrushao1994 @mbrookhart @sfvaroglu