Fix: resolve hash collisions by switching to UUID &correct logic for always-true statements & Update GPT api integration & Support qianwen-deepresearch (#10208)

### What problem does this PR solve?

Fix: resolve hash collisions by switching to UUID &correct logic for
always-true statements, solved: #10165
Feat: Update GPT api integration, solved: #10204 
Feat: Support qianwen-deepresearch, solved: #10163 
### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] New Feature (non-breaking change which adds functionality)

rag/llm/cv_model.py

@@ -75,7 +75,7 @@ class Base(ABC):
 
     def chat(self, system, history, gen_conf, images=[], **kwargs):
         try:
-            response = self.client.chat.completions.create(
+            response = self.client.responses.create(
                 model=self.model_name,
                 messages=self._form_history(system, history, images)
             )
@@ -87,7 +87,7 @@ class Base(ABC):
         ans = ""
         tk_count = 0
         try:
-            response = self.client.chat.completions.create(
+            response = self.client.responses.create(
                 model=self.model_name,
                 messages=self._form_history(system, history, images),
                 stream=True
@@ -174,7 +174,8 @@ class GptV4(Base):
 
     def describe(self, image):
         b64 = self.image2base64(image)
-        res = self.client.chat.completions.create(
+        # Check if this is a GPT-5 model and use responses.create API
+        res = self.client.responses.create(
             model=self.model_name,
             messages=self.prompt(b64),
         )
@@ -182,7 +183,7 @@ class GptV4(Base):
 
     def describe_with_prompt(self, image, prompt=None):
         b64 = self.image2base64(image)
-        res = self.client.chat.completions.create(
+        res = self.client.responses.create(
             model=self.model_name,
             messages=self.vision_llm_prompt(b64, prompt),
         )