[tiny-agents] Configure inference API key from inputs

[hanouticelina]

Same PR as huggingface/huggingface_hub#3226.

with this PR, the apiKey will be requested from the user via a prompt if not set as an environment variable, this was already implemented for server-specific credentials and env variables, this PR extends this for high-level credentials configuration. a config like this should work:

{
    "model": "gpt-4o",
    "provider": "openai",
    "apiKey": "${input:openai-api-key}",
    "inputs": [
		{
			"type": "promptString",
			"id": "hf-token",
			"description": "Your Hugging Face Token",
			"password": true
		},
		{
            "type": "promptString",
            "id": "openai-api-key",
            "description": "OpenAI API Key",
            "password": true
        }
		
	],
	"servers": [
		{
			"type": "http",
			"url": "https://huggingface.co/mcp",
			"headers": {
				"Authorization": "Bearer ${input:hf-token}"
			}
		}
	]
}

[Wauplin]

Thanks!