Feat: add BedrockCV for vision/image2text inference via LiteLLM (#14705)

## Summary

- `CvModel["Bedrock"]` was absent from `rag/llm/cv_model.py`, causing
`model_instance()` to return `None` when a Bedrock model was used as a
PDF parser — even after correct model resolution.
- This PR adds `BedrockCV`, enabling Bedrock vision models (e.g.
`amazon.nova-pro-v1:0`, `anthropic.claude-3-5-sonnet`) to be used as PDF
parsers.

## What problem does this PR solve?

When a Bedrock model is selected as the PDF parser in a knowledge base,
ingestion failed with:

```
'LiteLLMBase' object has no attribute 'describe_with_prompt'
```

The root cause: `LiteLLMBase` (the Bedrock chat implementation) was the
only registered handler for the Bedrock factory. It does not implement
`describe_with_prompt`. `CvModel` had no Bedrock entry, so
`model_instance()` returned `None` for `image2text` requests.

## Type of change

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

## Changes

**`rag/llm/cv_model.py`**

Adds `BedrockCV(Base)` with `_FACTORY_NAME = "Bedrock"`:

- Uses `litellm.completion` with the `bedrock/` prefix (consistent with
`LiteLLMBase`)
- Parses AWS credentials from the JSON key assembled by `add_llm`
(`auth_mode`, `bedrock_ak`, `bedrock_sk`, `bedrock_region`,
`aws_role_arn`)
- Supports three auth modes: `access_key_secret`, `iam_role` (via STS
`assume_role`), and default credential chain (IRSA, instance profile)
- Implements `describe_with_prompt` and `describe`

## Test plan

- [ ] Configure a Bedrock vision model (e.g. `amazon.nova-pro-v1:0`)
with valid AWS credentials
- [ ] Select it as PDF parser in a knowledge base
- [ ] Verify ingestion of a PDF document completes without errors
- [ ] Verify `CvModel["Bedrock"]` resolves to `BedrockCV`

🤖 Generated with [Claude Code](https://claude.ai/claude-code)

---------

Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

rag/llm/cv_model.py

@@ -1276,14 +1276,67 @@ class RAGconCV(GptV4):
     _FACTORY_NAME = "RAGcon"
     
     def __init__(self, key, model_name, lang="Chinese", base_url="", **kwargs):
-        
+
         if not base_url:
             base_url = "https://connect.ragcon.com/v1"
-        
+
         # Initialize client
         self.client = OpenAI(api_key=key, base_url=base_url)
         self.async_client = AsyncOpenAI(api_key=key, base_url=base_url)
         self.model_name = model_name
         self.lang = lang
-        
-        Base.__init__(self, **kwargs)
+
+        Base.__init__(self, **kwargs)
+
+
+class BedrockCV(Base):
+    _FACTORY_NAME = "Bedrock"
+
+    def __init__(self, key, model_name, lang="Chinese", **kwargs):
+        self.model_name = f"bedrock/{model_name}"
+        self.lang = lang
+        self._parse_credentials(key)
+        Base.__init__(self, **kwargs)
+
+    def _parse_credentials(self, key):
+        bedrock_key = json.loads(key)
+        self.auth_mode = bedrock_key.get("auth_mode", "")
+        self.aws_region = bedrock_key.get("bedrock_region", "us-east-1")
+        self.aws_ak = bedrock_key.get("bedrock_ak", "")
+        self.aws_sk = bedrock_key.get("bedrock_sk", "")
+        self.aws_role_arn = bedrock_key.get("aws_role_arn", "")
+
+    def _get_aws_creds(self):
+        if self.auth_mode == "access_key_secret":
+            return {
+                "aws_region_name": self.aws_region,
+                "aws_access_key_id": self.aws_ak,
+                "aws_secret_access_key": self.aws_sk,
+            }
+        elif self.auth_mode == "iam_role":
+            import boto3
+            sts_client = boto3.client("sts", region_name=self.aws_region)
+            resp = sts_client.assume_role(RoleArn=self.aws_role_arn, RoleSessionName="BedrockCVSession")
+            creds = resp["Credentials"]
+            return {
+                "aws_region_name": self.aws_region,
+                "aws_access_key_id": creds["AccessKeyId"],
+                "aws_secret_access_key": creds["SecretAccessKey"],
+                "aws_session_token": creds["SessionToken"],
+            }
+        else:
+            return {"aws_region_name": self.aws_region}
+
+    def describe_with_prompt(self, image, prompt=None):
+        import litellm
+        b64 = self.image2base64(image)
+        messages = self.vision_llm_prompt(b64, prompt)
+        res = litellm.completion(
+            model=self.model_name,
+            messages=messages,
+            **self._get_aws_creds(),
+        )
+        return res.choices[0].message.content.strip(), total_token_count_from_response(res)
+
+    def describe(self, image):
+        return self.describe_with_prompt(image)