Adds a schemas directory and one schema so far, for the OCM Discovery response in section 5

[lisad]

I'm not sure whether these are going to be normative or not, but I worked up a schema for section 5 to start with

[glpatcern]

Hi Lisa, great to see you starting on this, but it seems you may have not committed the schema itself or did I miss something?

[lisad]

I put a file called "ocm-discovery.json" inside a new folder called "schemas"? there will be several separate schemas if I continue with the rest of the spec. I made the schema a separate file for a number of reasons.

1. it's easier for us not to screw up the syntax when editing it (trying to editing valid JSON inside MD is error-prone for me), we can even add a github test action to confirm schemas are valid
2. folks will eventually be able to refer directly to these schemas when checking their implementations - some folks do this in CI/CD tests
3. I wasn't sure we'd want to have them in the spec, but if we do, we can still import the schema into the spec - see for example how Rohan Mahy uses {::include} within https://github.com/rohanmahy/mimi-message-status/blob/main/draft-mahy-mimi-message-status.md to merge the schemas into the draft.

[MahdiBaghbani]

@lisad I can see the file, but its contents are empty. Like Giuseppe, I assumed it hadn't been committed yet

[lisad]

Oh sorry I get it now! Fixed.

[glpatcern]

This seems a very good direction for all the schemas! As a matter of fact, currently the schemas are declared in the OpenAPI spec.yaml file, but that's not linked in the Draft, and instead we can both link them and enforce them in CIs.