perf: optional orjson JSON codec behind ENABLE_ORJSON (#27583)

Swap the JSON encoder/decoder used across the backend from stdlib json to
orjson when ENABLE_ORJSON is set — HTTP request bodies, JSONResponse
bodies, upstream provider responses, SSE chunks, and socket.io/Redis
payloads.

The flag defaults to off, in which case the app uses stdlib json and
engineio's codec verbatim, so default behaviour is unchanged.

- json_codec exports JSONCodec (stdlib json or the orjson codec) and
  SOCKETIO_JSON (engineio's codec or the orjson codec); call sites import
  JSONCodec and stay implementation-agnostic
- apply_orjson_http_json() is a no-op when the flag is off, leaving
  starlette's Request.json / JSONResponse.render untouched
- the orjson codec falls back to the stdlib for inputs orjson rejects
  (non-str dict keys, ints beyond 64 bits, NaN literals)
- orjson is imported only when the flag is on
- FastAPI(default_response_class=...) is deliberately not used: an
  explicit default disables the Pydantic direct-to-bytes fast path for
  response_model routes

backend/open_webui/utils/response.py

@@ -2,6 +2,7 @@ import json
 from numbers import Number
 from uuid import uuid4
 
+from open_webui.utils.json_codec import JSONCodec
 from open_webui.utils.misc import (
     openai_chat_chunk_message_template,
     openai_chat_completion_message_template,
@@ -237,7 +238,7 @@ async def convert_streaming_response_ollama_to_openai(ollama_streaming_response)
     completion_id = f'chatcmpl-{str(uuid4())}'
     first = True
     async for data in ollama_streaming_response.body_iterator:
-        data = json.loads(data)
+        data = JSONCodec.loads(data)
 
         model = data.get('model', 'ollama')
         message = data.get('message') or {}
@@ -268,7 +269,7 @@ async def convert_streaming_response_ollama_to_openai(ollama_streaming_response)
         if done and has_tool_calls:
             data['choices'][0]['finish_reason'] = 'tool_calls'
 
-        line = f'data: {json.dumps(data)}\n\n'
+        line = f'data: {JSONCodec.dumps(data)}\n\n'
         yield line
 
     yield 'data: [DONE]\n\n'