Using JUnit reporter (still) generates the wrong name per test

[SamLimpic]

In response to Fix JUnit reporter classname: this does not resolve the issue introduced in 3231, in response to Issue 3230

For instance this file contains 2 nested test directories:

{
  "info": {
    "_postman_id": "123",
    "name": "Unique Collection"
  },
  "item": [
    {
      "name": "Positive Test Cases",
      "item": [
        {
          "name": "Features",
          "item": [
            {
              "name": "Request",
              "event": [
                {
                  "listen": "test",
                  "script": {
                    "exec": [
                      "pm.test(`Testing unique Feature`, () => {});"
                    ]
                  }
                }
              ]
            }
          ]
        }
      ]
    },
    {
      "name": "Negative Test Cases",
      "item": [
        {
          "name": "Validations",
          "item": [
            {
              "name": "Request",
              "event": [
                {
                  "listen": "test",
                  "script": {
                    "exec": [
                      "pm.test(`Testing unique Validation`, () => {});"
                    ]
                  }
                }
              ]
            }
          ]
        }
      ]
    }
  ]
}

This single Unique Collection contains 2 tests, following this directory hierarchy:

PositiveTestCases > Features > Request --> "Testing unique Feature"
NegativeTestCases > Validations > Request --> "Testing unique Validation"

I would expect the JUnit report to output these tests on the report, as well as the collection the tests came from:

UniqueCollection.Testing unique Feature
UniqueCollection.Testing unique Validation

Instead, I get this:

PositiveTestCasesFeaturesRequest.Testing unique Feature
NegativeTestCasesValidationsRequest.Testing unique Validation

When multiple collections are run, they no longer display their respective Collection in the runner
Instead they list a series of directories without separation or punctuation, ending with the name of the failing request & test

This is much less helpful from a diagnostic perspective, where dozens of results for multiple collections are displayed simultaneously, with no additional indication of where each individual failure came from

PositiveTestCasesFeaturesRequest.Testing unique Feature
NegativeTestCasesValidationsRequest.Testing unique Validation
PositiveTestCasesFeaturesRequest.Testing unique Feature
NegativeTestCasesValidationsRequest.Testing unique Validation
PositiveTestCasesFeaturesRequest.Testing unique Feature
NegativeTestCasesValidationsRequest.Testing unique Validation
PositiveTestCasesFeaturesRequest.Testing unique Feature
NegativeTestCasesValidationsRequest.Testing unique Validation

vs

UnitTests.Testing unique Feature
UnitTests.Testing unique Validation
IntegrationTests.Testing unique Feature
IntegrationTests.Testing unique Validation
RegressionTests.Testing unique Feature
RegressionTests.Testing unique Validation
AcceptanceTests.Testing unique Feature
AcceptanceTests.Testing unique Validation

I would guess that this is the updated attribs.name.value displayed in place of the original formatted collection.name, but the impact of this change seems to be farther reaching than intended