Reworked get_organ_types

[DerekFurstPitt]

reworked get_organ_types such that it is only called once during the initialization of schema_manager. Created a global variable inside schema triggers and a quick setter function inside app.py to set that value after initialization of schema manager but before the value is fetched in schema_triggers

[yuanzhou]

@DerekFurstPitt Your rework looks be an anti-pattern. As you can tell we don't use any top-level variables in schema_triggers.py, it contains only functions. And schema_manager.py uses global variables but it only exposes them via getter functions, schema_manager.get_organ_types() in this case.

Instead of working around circular import, there's a much simpler solution. Take a look at how auth_helper_instance = get_auth_helper_instance() is being used in schema_manager.py first. Then

• Add this new _organ_types = None
• In get_organ_types(), first check if _organ_types is available then just return it. Otherwise set its value to be the result of the first response. This is essentially a singleton pattern.
• In schema_triggers.py and app.py, keep using the old schema_manager.get_organ_types() since now it just reuses the value.
• In app.py, there's no need to use the second parameter for validate_organ_code(organ_code, organ_types_dict=None).

Keep in mind that we are using the schema_manager as the central piece, any values to be reused by app.py or schema_triggers.py just need a reference/pointer to the exposed function in schema_manager.