Improve configs - `ModelConfig`

[hmellor]

Follows #16332, #16422, #16533, #16603, #16789, #16835, #16971, #16980

Notable changes:

• optional_x has been replaced with optional_type(x) so we don't need to keep adding new methods for new types
• Type hints on QuantizationConfig and LLM classes have been made more specific, using a Literal of possible options instead of str
• The compressed tensors quantization method was referred to as compressed-tensors and compressed_tensors in different places, they've now been standardised to compressed_tensors
• Quantization method overrides are more clearly commented and enforced in config.py
• ModelConfig is no longer used as a key to a dictionary in the multimodal registry (keys are usually simple and hashable, a large dataclass is neither of those things)

[github-actions[bot]]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

🚀

[hmellor]

I think I need a better solution to the hashing error. If possible, it would be better not to need ModelConfig to be hashable at all.

[mergify[bot]]

This pull request has merge conflicts that must be resolved before it can be merged. Please rebase the PR, @hmellor.

https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks/syncing-a-fork

[hmellor]

Since this PR has become quite big, I've been splitting it up. You can see the description for the sub-PRs.