Ensure prompt validation when attaching prompts to generations

[alex-stoica]

Describe the feature or potential improvement

Currently, it is possible to attach any existing prompt to a generation without proper validation. This can lead to scenarios where the prompt used for the generation does not match the attached prompt, causing inconsistencies in the process.

Please ensure that prompt integrity is maintained when adding a prompt to a generation process.

Additional information

Example scenario:

try:
    prompt = tracer.get_prompt(prompt_name)
    if prompt:
        span.update(prompt=prompt)  # langfuse.client.StatefulSpanClient
except Exception as e:
    print(f"Prompt not found or error occurred: {e}")