Order fileMatches from more specific to more generic

I have a .github/workflows/site.yml file that matches multiple entries from schemastore catalog:

    {
      "name": "Ansible Playbook",
      "description": "Ansible playbook files",
      "fileMatch": [
        "playbook.yml",
        "playbook.yaml",
        "site.yml",
        "site.yaml",
        "**/playbooks/*.yml",
        "**/playbooks/*.yaml"
      ],
      "url": "https://raw.githubusercontent.com/ansible/ansible-lint/main/src/ansiblelint/schemas/ansible.json#/$defs/playbook"
    }

and

    {
      "name": "GitHub Workflow",
      "description": "YAML GitHub Workflow",
      "fileMatch": [
        "**/.github/workflows/*.yml",
        "**/.github/workflows/*.yaml",
        "**/.gitea/workflows/*.yml",
        "**/.gitea/workflows/*.yaml",
        "**/.forgejo/workflows/*.yml",
        "**/.forgejo/workflows/*.yaml"
      ],
      "url": "https://www.schemastore.org/github-workflow.json"
    }

Unfortunately, actually, Ansible Playbook is selected, which generates lots of false warnings in the file.

Could it be possible to prioritize the selection of the schema according to the number of characters matched outside of wildcards?
This way the more specific fileMatch will be selected if more that one match.


For example in my case, .github/workflows/site.yml:

• matches 8 characters for site.yml
• matches 23 characters for **/.github/workflows/*.yml, excluding characters matched by wildcards

The second one is more specific characters and should be preferred


This should fix most of the cases relatively easily
Regards

Start by tracing how the language service selects among matching schema fileMatch entries for .github/workflows/site.yml; compare the literal portions of the Ansible and GitHub Workflow patterns shown in the issue. Done when the more-specific GitHub Workflow schema is selected for this example and the behavior is covered by an appropriate regression test.