Fix error of changing embedding model (#4184)

### What problem does this PR solve?

1. Change embedding model of knowledge base won't change the default
embedding model.
2. Retrieval test bug

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

---------

Signed-off-by: jinhai <haijin.chn@gmail.com>

rag/llm/embedding_model.py

@@ -61,11 +61,11 @@ class DefaultEmbedding(Base):
         ^_-
 
         """
-        if not settings.LIGHTEN and not DefaultEmbedding._model:
+        if not settings.LIGHTEN:
             with DefaultEmbedding._model_lock:
                 from FlagEmbedding import FlagModel
                 import torch
-                if not DefaultEmbedding._model:
+                if not DefaultEmbedding._model or model_name != DefaultEmbedding._model_name:
                     try:
                         DefaultEmbedding._model = FlagModel(os.path.join(get_home_cache_dir(), re.sub(r"^[a-zA-Z0-9]+/", "", model_name)),
                                                             query_instruction_for_retrieval="为这个句子生成表示以用于检索相关文章：",
@@ -261,7 +261,7 @@ class FastEmbed(DefaultEmbedding):
             threads: int | None = None,
             **kwargs,
     ):
-        if not settings.LIGHTEN and not FastEmbed._model:
+        if not settings.LIGHTEN:
             with FastEmbed._model_lock:
                 from fastembed import TextEmbedding
                 if not DefaultEmbedding._model or model_name != DefaultEmbedding._model_name: