Add OpenAI latest models to TokenValidator

[kokuyouwind]

Description

fix #594

Based on OpenAI's most recent list of models, the missing models were added and updates to existing models were reflected.
I checked the specs for OpenAIValidator and I don't think it needs to be updated.

I ran the code in the original Issue and confirmed that the error no longer occurs.

> llm = Langchain::LLM::OpenAI.new(
   api_key: ENV.fetch('OPENAI_API_KEY', nil),
      default_options: {
        chat_completion_model_name: 'gpt-4-turbo'
    })
> llm.chat(messages: [{ role: 'user', content: 'hello'}])
=>
#<Langchain::LLM::OpenAIResponse:0x000000012b239340
 @model=nil,
 @raw_response=
  {"id"=>"chatcmpl-9IBSpkOQdaXUcax5Y58IVYEHKNvQd",
   "object"=>"chat.completion",
   "created"=>1714120999,
   "model"=>"gpt-4-turbo-2024-04-09",
   "choices"=>
    [{"index"=>0, "message"=>{"role"=>"assistant", "content"=>"Hello! How can I assist you today?"}, "logprobs"=>nil, "finish_reason"=>"stop"}],
   "usage"=>{"prompt_tokens"=>8, "completion_tokens"=>9, "total_tokens"=>17},
   "system_fingerprint"=>"fp_ea6eb70039"}>

[kokuyouwind]

📝
gpt-3.5-turbo is now aliased to gpt-3.5-turbo-0125, so I aligned it there.

[andreibondarev]

@kokuyouwind Do you find there's value in this token length validation? I've been thinking of disabling it by default.

[kokuyouwind]

@andreibondarev

@kokuyouwind Do you find there's value in this token length validation? I've been thinking of disabling it by default.

I have created this PR with respect to the existing implementation, but I too think this validation is unnecessary.
If there is a problem with the number of tokens, the OpenAI API will return an error, and I don't think there is a particular need to save the number of API runs.