Update some type hints

[zucchini-nlp]

An eos token can also be a list on most recent models, so this PR allows all EOS in config be a list as well. Same for q-lora-rank which apparently can be an explicit None for some model

Also bring back layer_type_validation and add deprecation message. I realized it's actually used a lot and will break remote code as well

WIP, will collect more type hint fixes during the day from vLLM CI jobs

[github-actions]

[For maintainers] Suggested jobs to run (before merge)

run-slow: blenderbot, blenderbot_small, clvp, deepseek_v3, dia, eurobert, exaone_moe, flaubert, fnet, fsmt, git, glm4_moe_lite, gpt2, gpt_bigcode, gpt_neo, gptj

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.