Null values, rate limit fix and gpt-4o

[nugaon]

I had problem with running the built executable (since it is necessary to build, see #1286) so I made some changes in the source code that may fix issues for those who want to continue using this awesome application.

Small changes but includes:

• fixing openai requests ended up in 400 Bad Request
• trying again requests failed with 429 Rate Limitation after 5 seconds delay
• default model changed to gpt-4o instead of gpt-4-Turbo which is 2 times faster on half price
• logging sent messages to see how Bleep communicates with OpenAI : )