Large _pydantic shared folder

Currently any function which handles any sort of models and validation which is most if not all of them will include the whole `_pydnantic` shared directory. This includes all of the models for all of the functions. At the moment it is fine. But as the number of endpoints and therefore models grows it might become too big.

It might be better to define in the function definition the patter which pertains to that functions models. Plus the shared pydantic utils such as dynamodb and the extended models. With the possibility of splitting the extended models by function later.

The shared models (models used across multiple functions) can be labelled with a tag in the `sls.{fun}.models.yml` definition of the model and it will be added to `models_shared.py` which can be added to anyone who needs to use models across functions.

An example of this would be a ticket will need to access the Customer model but not necessarily the CreateCustomerRequest model or DeleteCustomerRequest model.