feature: define beta helper methods for language models

[overtonch]

Before this PR

Currently, calling language models from Python Functions is possible but requires manually setting up URLs and reading from context vars directly.

Working code snippet today:

from functions.api import String, get_api_gateway_base_url, function
from foundry_sdk_runtime.context_vars import FOUNDRY_TOKEN
from openai import OpenAI

@function
def call_openai(prompt: String) -> String:
    token = FOUNDRY_TOKEN.get()
    api_base_url = get_api_gateway_base_url() + "/api/v2/llm/proxy/openai/v1"
    client = OpenAI(
        api_key=token,
        base_url=api_base_url,
    )

    completion = client.chat.completions.create(
        model="gpt-5",
        messages=[
            {
                "role": "user",
                "content": prompt,
            },
        ],
    )

    return str(completion.choices[0].message.content)

After this PR

==COMMIT_MSG==
Beta util methods are exposed to enable easily calling language models using the proxy endpoint.
==COMMIT_MSG==

Example code using the utils from this PR:

  from functions.api import String, function
  from foundry_sdk.v2.functions import get_foundry_token, get_openai_base_url, get_http_client
  from openai import OpenAI

  @function
  def call_openai(prompt: String) -> String:
      client = OpenAI(
          api_key=get_foundry_token(preview=True),
          base_url=get_openai_base_url(preview=True),
          http_client=get_http_client(preview=True),
      )

      completion = client.chat.completions.create(
          model="gpt-5",
          messages=[{"role": "user", "content": prompt}],
      )

      return str(completion.choices[0].message.content)

Possible downsides?

[changelog-app]

✅ Successfully generated changelog entry!

Entry generated via PR title

To modify this entry, edit PR title using proper format.

---

📋Changelog Preview

✨ Features

• Define beta helper methods for language models (#320)