[release] Add GitHubClient: pull request support

[andrew-anyscale]

Add GitHubPullUser, GitHubPull dataclasses and GitHubRepo.get_pull().
Tests cover successful fetch, 404 error, and auth header verification.

Topic: github-client-pulls
Relative: github-client
Signed-off-by: andrew andrew@anyscale.com

[andrew-anyscale]

Reviews in this chain:
└#61767 [release] Add GitHubClient: pull request support
　└#61768 [release] Add GitHubClient: issue support
　　└#61710 [release] Replace PyGithub with GitHubClient in ray_release
　　　└#61711 [ci] Replace PyGithub with GitHubClient in get_contributors.py
　　　　└#61770 [release] Add --no-upgrade to pip-compile args
　　　　　└#61712 [release] Remove PyGithub from requirements

[andrew-anyscale]

#	head	base	diff	date	summary
0	0a40bf83	bbd8bef0	diff	Mar 16 11:51 AM	2 files changed, 78 insertions(+)
1	40ba2ee5	bbd8bef0	diff	Mar 16 12:09 PM	1 file changed, 9 insertions(+), 4 deletions(-)
2	d7447e9a	7a62c7fd	rebase	Mar 16 12:47 PM	0 files changed
3	ae5a4a0e	0ce9d7ae	rebase	Mar 16 15:17 PM	0 files changed
4	75b1f4ba	b86462c1	diff	Mar 16 17:22 PM	0 files changed

[gemini-code-assist]

Code Review

This pull request adds functionality to fetch a single pull request from GitHub by its number. The changes include new dataclasses for the pull request and its author, a new method in GitHubRepo to perform the fetch, and comprehensive tests covering the happy path, error cases, and authentication headers. The implementation is clear and well-tested. I have one suggestion to improve the code structure by encapsulating the object creation logic within the dataclasses themselves, which will enhance maintainability.

[gemini-code-assist]

For better separation of concerns and maintainability, consider moving the logic for creating GitHubPull and GitHubPullUser instances from the API response dictionary into factory class methods on the dataclasses themselves.

This would look something like this:

@dataclass
class GitHubPullUser:
    login: str

    @classmethod
    def from_dict(cls, data: dict) -> "GitHubPullUser":
        return cls(login=data["login"])

@dataclass
class GitHubPull:
    number: int
    user: GitHubPullUser

    @classmethod
    def from_dict(cls, data: dict) -> "GitHubPull":
        return cls(
            number=data["number"],
            user=GitHubPullUser.from_dict(data["user"])
        )

# In GitHubRepo:
def get_pull(self, number: int) -> GitHubPull:
    """Return a single pull request by number."""
    data = self._client._get(f"/repos/{self.full_name}/pulls/{number}")
    return GitHubPull.from_dict(data)

This pattern makes your dataclasses responsible for their own construction from the raw API data, which is a cleaner design that will be easier to maintain and extend.

[aslonnie]

this test fixture is really neat~