RequstReply: reply channel will not be displayed

[GreenRover]

Description

When an operation contains an reply channel

operations:
  requestReply:
    action: send
    channel:
      $ref: '#/channels/request'
    reply:
      channel:
        $ref: '#/channels/response'

In react ui the request channel will be displayed instead of the reply channel.

Reproducer

asyncapi: 3.0.0
info:
  title: Request reply example api
  version: '1.0.0'
  description: |
    THIS IS NOT A STANDALONE API!!!
    DONT COPY PASTE!!!
    Please see this as a template that should be adopted and integrated into your api.
    Because this service is part of your application.
channels:
  request:
    address: tms/iad/lta/p/v1/prod/my/request
    description: Get a compacted view of the past.
    messages:
      request:
        $ref: '#/components/messages/request'
  response:
    address: null
    description: The response to your request
    messages:
      normalResponse:
        $ref: '#/components/messages/normalResponse'
      groupedResponse:
        $ref: '#/components/messages/groupedResponse'

operations:
  requestReply:
    action: send
    channel:
      $ref: '#/channels/request'
    reply:
      channel:
        $ref: '#/channels/response'
      messages:
        - $ref: '#/channels/response/messages/normalResponse'
        - $ref: '#/channels/response/messages/groupedResponse'
      address:
        location: '$message.header#/replyTo'
    description: >
      Request reply example operation

components:
  messageTraits:
    response:
      headers:
        type: object
        required:
          - totalReplies
          - correlationId
        properties:
          totalReplies:
            type: number
            minimum: -1
            description: |
              The total number of reply you have to expect.
              
              Edge cases:
              -  "0" There are no responses and/or there is an error. The error case indicates end of stream.
                     If `errorMessage` is not empty.  Your requests could not be processed.
                     If `errorMessage` is empty or not set, your requests dint find anything to return aka empty array.
              - "-1" The number of replies is unknown.
                     In this edge case only the last message will have a totalReplies >= 1 but the body of the EndOfStream
                     message will not be evaluated.
          correlationId:
            type: string
            description: Correlation ID set by application. To identify response to request
          errorMessage:
            type: string
            description: |
              In case of that the request was invalid. This field is given with an error message and totalReplies is set to 0
      correlationId:
        location: $message.header#/correlationId
  messages:
    request:
      name: request
      title: Your question
      headers:
        type: object
        additionalProperties: false
        required:
          - replyTo
          - correlationId
        properties:
          replyTo:
            type: string
            pattern: ^tms/iad/lta/p-\w+/v1/prod/your/reply/([\w\-]{10,})$
            maxLength: 134
            description: |
              the address where you receive the reply to.
              Use the solace ReplyTo header:
              https://docs.solace.com/API-Developer-Online-Ref-Documentation/java/com/solacesystems/jcsmp/XMLMessage.html#getReplyTo--
              
              TODO: you need to adopt the pattern to match your topic from configuration.
              
              But be aware of to:
              - use a very unique system identifier. To avoid collisions and debugging purpose to identify you.
              - but not exceed the 134 char max topic length
            externalDocs:
              url: https://confluence.sbb.ch/x/Ohbleg
              description: documentation for request reply handling at tms
          groupedMessages:
            type: boolean
            description: |
              Indicates that at a multi response channel a grouped messages is acceptable.
              This flag should be `true`if:
              - You expect multiple response > 200
              - You expect small to very small response messages < 100kb
              - You dont process the header of the original update messages. Because on grouped messages all header get lost.
              
              Benefit of grouped messages:
              - The overhead will be minimized 
              - You can receive much faster
          correlationId:
            type: string
            description: Correlation ID set by application. To identify response to request
      correlationId:
        location: $message.header#/correlationId


    normalResponse:
      name: reply
      title: Reply to a request
      payload:
        type: object
        description: |
          TODO: Define here the payload Schema.
          Transfer encoded as byte array.
      headers:
        type: object
        required:
          - replyIndex
        properties:
          replyIndex:
            type: number
            minimum: 0
            description: The index of the reply message
      traits:
        - $ref: "#/components/messageTraits/response"

    groupedResponse:
      name: replyGrouped
      title: Grouped reply to a request
      description: |
        Payload contains multiple small responses in a list.
        Should be used to reduce the overhead in case a response consists of many (>200) small messages (<100kb).
      payload:
        type: array
        items:
          oneOff:
            - type: string
              enum:
                - TextMessage
                - XMLContentMessage
                - BytesMessage
                - MapMessage
                - StreamMessage
              description: The message type
            - type: object
              description: |
                TODO: Define here the payload Schema.
                Transfer encoded as byte array.
        description: |
          A list of multiple message bodies. 
          The content is a alternating list with: MessageType, body, MessageType, body, ...
          Needs to be of type solace "StreamMessage"
      headers:
        type: object
        required:
          - replyIndex
          - groupedMessages
        properties:
          replyIndex:
            type: string
            description: |
              The index range of the reply message.
            examples:
              - 0
              - 0-3
              - 4-25
          groupedMessages:
            type: boolean
            description: needs to be `true` to indicate a grouped "groupedResponse" message

      traits:
        - $ref: "#/components/messageTraits/response"

[derberg]

fixed with #856