Internal server error for role orders in LLM inference

Hi,

I am getting 500s with description

chat messages must alternate roles between 'user' and 'assistant'.  Message may have a leading 'system' role message

and

"Internal Server Error\",\"status\":500,\"detail\":\"Last message role should be 'user'

I think this order validation is unnecessary. Most of the models (e.g. mixtrals, llama3, gemma etc) are perfectly fine with any order of roles. This claim can be validated on groq's playground: https://console.groq.com/playground

This issue currently breaks some existing patterns, like continuations (without the user explicitly saying "continue") or in some cases running agents with observations etc.

首先跟踪产生所述角色顺序和最后一条消息错误的验证逻辑，然后复现一个包含连续消息或代理观察消息的请求。当受支持的非交替角色序列不再返回 HTTP 500，并且预期的验证行为仍然得到覆盖时，即表示完成。