Spike: Review semi structured data mechanism

Background context To review the mechanism to understand if we can make it writable. We still need to retain control of which fields are accessible so having an endpoint to surface all semi structured data is really a non-starter as some of the information could be confidential that needs to remain protected. As a result, the existing outward contract remains the same.

Specification

Project: Reapit.Services.Properties / Reapit.Packages.ValueObject (this is where the SemiStructured class lives)
We need to know what the data type of each field is, and have control over whether or not it should be writeable. Therefore, consider the following changes to the configuration object for each customer:

"extrasConfig": {
  "properties": {
   "pSituation": {
       "type": "string|bool|datetime",
       "writeable": true|false
   }
  } 
}

If a field is not writeable and somebody includes the information in a PATCH, it should result in a validation error
If a field is writeable but is provided in the wrong data type (eg string in a date field), it should result in a validation error
If validation succeeds, call SetExtra() to update the value accordingly.
This may also require a change to the PlatformIdentity object where the configuration gets pulled in from.

reapit / foundations

Spike: Review semi structured data mechanism #10572