chore: Add OpenAI compatibility for Ollama embeddings (#2440)

# What does this PR do?
This PR adds OpenAI compatibility for Ollama embeddings. Closes
https://github.com/meta-llama/llama-stack/issues/2428

Summary of changes:
- `llama_stack/providers/remote/inference/ollama/ollama.py`
- Implements the OpenAI embeddings endpoint for Ollama, replacing the
NotImplementedError with a full function that validates the model,
prepares parameters, calls the client, encodes embedding data
(optionally in base64), and returns a correctly structured response.
- Updates import statements to include the new embedding response
utilities.

- `llama_stack/providers/utils/inference/litellm_openai_mixin.py`
- Refactors the embedding data encoding logic to use a new shared
utility (`b64_encode_openai_embeddings_response`) instead of inline
base64 encoding and packing logic.
   - Cleans up imports accordingly.

- `llama_stack/providers/utils/inference/openai_compat.py`
- Adds `b64_encode_openai_embeddings_response` to handle encoding OpenAI
embedding outputs (including base64 support) in a reusable way.
- Adds `prepare_openai_embeddings_params` utility for standardizing
embedding parameter preparation.
   - Updates imports to include the new embedding data class.

- `tests/integration/inference/test_openai_embeddings.py`
- Removes `"remote::ollama"` from the list of providers that skip OpenAI
embeddings tests, since support is now implemented.

## Note

There was one minor issue, which required me to override the
`OpenAIEmbeddingsResponse.model` name with
`self._get_model(model).identifier` name, which is very unsatisfying.

## Test Plan
Unit Tests and integration tests

---------

Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>

llama_stack/providers/utils/inference/openai_compat.py

@@ -3,8 +3,10 @@
 #
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
+import base64
 import json
 import logging
+import struct
 import time
 import uuid
 import warnings
@@ -108,6 +110,7 @@ from llama_stack.apis.inference.inference import (
     OpenAIChatCompletion,
     OpenAICompletion,
     OpenAICompletionChoice,
+    OpenAIEmbeddingData,
     OpenAIMessageParam,
     OpenAIResponseFormatParam,
     ToolConfig,
@@ -1483,3 +1486,55 @@ class OpenAIChatCompletionToLlamaStackMixin:
             model=model,
             object="chat.completion",
         )
+
+
+def prepare_openai_embeddings_params(
+    model: str,
+    input: str | list[str],
+    encoding_format: str | None = "float",
+    dimensions: int | None = None,
+    user: str | None = None,
+):
+    if model is None:
+        raise ValueError("Model must be provided for embeddings")
+
+    input_list = [input] if isinstance(input, str) else input
+
+    params: dict[str, Any] = {
+        "model": model,
+        "input": input_list,
+    }
+
+    if encoding_format is not None:
+        params["encoding_format"] = encoding_format
+    if dimensions is not None:
+        params["dimensions"] = dimensions
+    if user is not None:
+        params["user"] = user
+
+    return params
+
+
+def b64_encode_openai_embeddings_response(
+    response_data: dict, encoding_format: str | None = "float"
+) -> list[OpenAIEmbeddingData]:
+    """
+    Process the OpenAI embeddings response to encode the embeddings in base64 format if specified.
+    """
+    data = []
+    for i, embedding_data in enumerate(response_data):
+        if encoding_format == "base64":
+            byte_array = bytearray()
+            for embedding_value in embedding_data.embedding:
+                byte_array.extend(struct.pack("f", float(embedding_value)))
+
+            response_embedding = base64.b64encode(byte_array).decode("utf-8")
+        else:
+            response_embedding = embedding_data.embedding
+        data.append(
+            OpenAIEmbeddingData(
+                embedding=response_embedding,
+                index=i,
+            )
+        )
+    return data