[FEA]: Simplify GraphQL in our project automation

### Is this a duplicate?

- [X] I confirmed there appear to be no [duplicate issues](https://github.com/NVIDIA/cccl/issues) for this request and that I agree to the [Code of Conduct](CODE_OF_CONDUCT.md)

### Area

Infrastructure

### Is your feature request related to a problem? Please describe.

The automations are hard to read, partly because graphQL isn't a beautiful query language IMO, and partly because they're inefficient queries.

### Describe the solution you'd like

We can and should use the global node IDs of items and query them directly. In action below:

```
query {
organization(login: "${{ env.ORG }}") {
repository(name: "${{ env.REPO }}") {
issueOrPullRequest(number: ${{ env.PR_NUMBER }}) {
... on PullRequest {
id
projectItems(first: 10) {
edges {
node {
id
project {
id
}
}
}
}
}
}
}
}
}' > project_data.json
```

Can be simplified into:
```
query {
node(id: "${{ env.PR_NODE_ID }}") {
... on PullRequest {
projectItems(first: 10) {
nodes {
id
project {
id
}
}
}
}
}
}' > project_data.json
```

Similarly, we don't need separate steps for an issue or PR to get its project specific ID, we can just modify the above query to include `... on Issue` and query that. The output JSON will be the same regardless and downstream mutations don't care if it's a PR or an issue.

### Describe alternatives you've considered

_No response_

### Additional context

This saves us graphQL api usage, and it's always good to minimize that since we don't want to get rate limited on a busy day.