fix: SSE data reporting

[danielpeintner]

fixes #1294

[egekorkan]

Some comments:

• Here is a Test Thing you can use to test the code: https://github.com/eclipse-thingweb/test-things/blob/main/things/calculator/http/express/http-simple-calculator-thing.td.jsonld the code is in there as well. There are client tests with node-wot in the same folder but they are commented out until this is resolved
• There are some issue threads in the standards: SSE Default Method w3c/wot-binding-templates#330 , [http] Differentiating sse endpoint w3c/wot-binding-templates#342 , Which content-type to use for SSE? w3c/wot-binding-templates#347
• We had a chat with @relu91 on this and the form should be application/json and implicit knowledge of sending text/event-stream comes from the subprotocol. I am not exactly happy with this but this is what we can do now. With SSE, there is the possibility to have three media types at once:
  1. What is on the wire in the payload: text/event-stream . This is always the case and this must be set in the accept header in the first request
  2. What is in the data part of the payload: This can be just text, stringified JSON or any other stringified media type (we can have CBOR or XML in there too)
  3. What the Consumer application wants: With JSON Schema and JSON-based thinking we have now, this is always JS types for node-wot.

Given that 1 is always the case for sse and we can understand that from subprotocol, it is fine to put json in the form content type.

[danielpeintner]

Given that 1 is always the case for sse and we can understand that from subprotocol, it is fine to put json in the form content type.

This is something that (I think) goes beyond this PR. I suggest to open another issue/PR for that.

@egekorkan w.r.t. the other comments I am not sure whether they affect "this" PR. There is a test added that checks the consistency of the change. Do you want me to change the PR further?

[egekorkan]

Given that 1 is always the case for sse and we can understand that from subprotocol, it is fine to put json in the form content type.

This is something that (I think) goes beyond this PR. I suggest to open another issue/PR for that.

@egekorkan w.r.t. the other comments I am not sure whether they affect "this" PR. There is a test added that checks the consistency of the change. Do you want me to change the PR further?

I think I got confused after some discussions with @relu91 and commented here. My comments do not affect this PR.

[VigneshVSV]

works