Multilora docs

[natke]

Staged here:

• Tutorial: https://natke.github.io/onnxruntime/docs/genai/tutorials/finetune.html
• Python API docs: https://natke.github.io/onnxruntime/docs/genai/api/python.html#adapter-class
• C API docs: https://natke.github.io/onnxruntime/docs/genai/api/c.html#adapter-api
• C# API docs: https://natke.github.io/onnxruntime/docs/genai/api/csharp.html#adapter-class
• Java API docs: https://natke.github.io/onnxruntime/docs/genai/api/java.html#adapter-class ;-)
• Adapter file format reference: https://natke.github.io/onnxruntime/docs/genai/reference/adapter.html

[baijumeswani]

Could we instead point users to the schema inside the repo so we don't need to copy the contents every time there is a change?
https://github.com/microsoft/onnxruntime/blob/main/onnxruntime/lora/adapter_format/adapter_schema.fbs

[baijumeswani]

Suggested change

	`OgaResult` containing an error message if the adapter failed to unload. This can occur if the method is called with an adapter that is not already loaded.
	`OgaResult` containing an error message if the adapter failed to unload. This can occur if the method is called with an adapter that is not already loaded or has been marked active by a `OgaGenerator` still in use.

[baijumeswani]

While this is ok, shall we update model-qa.py in onnxruntime-genai to add support for loading adapters?