Add packages lock file support for aliased projects with duplicate frameworks

[nkolev92]

Bug

Fixes: NuGet/Home#14538

Description

This PR adds support for a V3 formatted lock file.
Unlike the assets file, this file is checked in, so we should not cause the lock file to upgrade unless absolutely necessary.
As such the v3 format will only be enabled when there are duplicate frameworks.

The only real contentious piece here is the format of the new lock file.

The spec is at NuGet/Home#14760.

The current lock file has a dependencies element and tfm/rid as a key.

{
  "version": 1,
  "dependencies": {
    "net10.0": {
      "Newtonsoft.Json": {
        "type": "Direct",
        "requested": "[13.0.4, )",
        "resolved": "13.0.4",
        "contentHash": "pdgNNMai3zv51W5aq268sujXUyx7SNdE2bj1wZcWjAQrKMFZV260lbqYop1d2GM67JI1huLRwxo9ZqnfF/lC6A=="
      }
    }
  }
}

The new variation is as below.
Note that the order of dependencies a alias/rid is actually being inverted to avoid adding redundant depth.

{
  "version": 3,
  "alias/rid": {
    "framework": "net10.0",
    "dependencies": {
      "Newtonsoft.Json": {
        "type": "Direct",
        "requested": "[13.0.4, )",
        "resolved": "13.0.4",
        "contentHash": "pdgNNMai3zv51W5aq268sujXUyx7SNdE2bj1wZcWjAQrKMFZV260lbqYop1d2GM67JI1huLRwxo9ZqnfF/lC6A=="
      }
    }
  }
}

You can read about the alternatives in the PR at NuGet/Home#14760

PR Checklist

• Meaningful title, helpful description and a linked NuGet/Home issue
• Added tests
• Link to an issue or pull request to update docs if this PR changes settings, environment variables, new feature, etc.

[Nirmal4G]

I vote for the last one, since alias is just an alias for the target framework! 🤔

{
  "version": 3,
  "alias/rid": {
    "framework": "net10.0",
    "dependencies": {
      "Newtonsoft.Json": {
        "type": "Direct",
        "requested": "[13.0.4, )",
        "resolved": "13.0.4",
        "contentHash": "pdgNNMai3zv51W5aq268sujXUyx7SNdE2bj1wZcWjAQrKMFZV260lbqYop1d2GM67JI1huLRwxo9ZqnfF/lC6A=="
      }
    }
  }
}

[nkolev92]

I've been having some internal discussions and that seems to be the consensus too :)