Confusing @model() syntax

In LB3, users could specify model settings at two levels: as a root property or inside options property.

See https://stackoverflow.com/q/53307168/69868 for an example of LB3 syntax applied in an LB4 project:

@model({
  settings: {strict: false},
  name: 'client',
  plural: 'clients',
  options: {
    mongodb: {
      collection: 'clients',
    },
  },
})
export class Client extends Entity {
  // ...
}

I am proposing to make two changes in LB4 to help users coming from LB3:

1. Recognize options the same way as settings. In the example above, mongodb settings are not picked by LB4 now. With the proposed change in place, LB4 will set collection to clients as expected. Alternatively, tell the user setting options that they are trying to set an unsupported model-definition property. This can be done at compiler level too.

2. Allow settings to be provided as top-level properties, for example:

   @model({
     name: 'client',
     strict: false,
     mongodb: {
       collection: 'clients',
     },
   })
   export class Client extends Entity {
     // ...
   }
   

Start at the @model() decorator and trace how LB4 reads model settings, comparing that behavior with the LB3 examples in the issue. Resolve whether options should be accepted or rejected and whether top-level settings should be supported; done means the chosen behavior is implemented and verified for the Client examples.