Feat: add gpustack model provider (#4469)

### What problem does this PR solve?

Add GPUStack as a new model provider.
[GPUStack](https://github.com/gpustack/gpustack) is an open-source GPU
cluster manager for running LLMs. Currently, locally deployed models in
GPUStack cannot integrate well with RAGFlow. GPUStack provides both
OpenAI compatible APIs (Models / Chat Completions / Embeddings /
Speech2Text / TTS) and other APIs like Rerank. We would like to use
GPUStack as a model provider in ragflow.

[GPUStack Docs](https://docs.gpustack.ai/latest/quickstart/)

Related issue: https://github.com/infiniflow/ragflow/issues/4064.

### Type of change

- [x] New Feature (non-breaking change which adds functionality)



### Testing Instructions
1. Install GPUStack and deploy the `llama-3.2-1b-instruct` llm, `bge-m3`
text embedding model, `bge-reranker-v2-m3` rerank model,
`faster-whisper-medium` Speech-to-Text model, `cosyvoice-300m-sft` in
GPUStack.
2. Add provider in ragflow settings.
3. Testing in ragflow.

rag/llm/rerank_model.py

@@ -18,10 +18,12 @@ import threading
 from urllib.parse import urljoin
 
 import requests
+import httpx
 from huggingface_hub import snapshot_download
 import os
 from abc import ABC
 import numpy as np
+from yarl import URL
 
 from api import settings
 from api.utils.file_utils import get_home_cache_dir
@@ -457,3 +459,53 @@ class QWenRerank(Base):
             return rank, resp.usage.total_tokens
         else:
             raise ValueError(f"Error calling QWenRerank model {self.model_name}: {resp.status_code} - {resp.text}")
+
+class GPUStackRerank(Base):
+    def __init__(
+            self, key, model_name, base_url
+    ):
+        if not base_url:
+            raise ValueError("url cannot be None")
+
+        self.model_name = model_name
+        self.base_url = str(URL(base_url)/ "v1" / "rerank")
+        self.headers = {
+            "accept": "application/json",
+            "content-type": "application/json",
+            "authorization": f"Bearer {key}",
+        }
+
+    def similarity(self, query: str, texts: list):
+        payload = {
+            "model": self.model_name,
+            "query": query,
+            "documents": texts,
+            "top_n": len(texts),
+        }
+
+        try:
+            response = requests.post(
+                self.base_url, json=payload, headers=self.headers
+            )
+            response.raise_for_status()
+            response_json = response.json()
+
+            rank = np.zeros(len(texts), dtype=float)
+            if "results" not in response_json:
+                return rank, 0
+
+            token_count = 0
+            for t in texts:
+                token_count += num_tokens_from_string(t)
+
+            for result in response_json["results"]:
+                rank[result["index"]] = result["relevance_score"]
+
+            return (
+                rank,
+                token_count,
+            )
+
+        except httpx.HTTPStatusError as e:
+            raise ValueError(f"Error calling GPUStackRerank model {self.model_name}: {e.response.status_code} - {e.response.text}")
+