feat: add BGE embedding model for concurrent operation with Llama

- Create config_bge.yaml for BAAI/bge-large-en-v1.5 on port 8002
- Reduce Llama VRAM to 70% and context to 16K for concurrent use
- Add BGE service to supervisor with vllm group

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

Co-Authored-By: Claude <noreply@anthropic.com>

supervisord.conf

@@ -74,6 +74,24 @@ priority=200
 stopwaitsecs=30
 
 
+# vLLM BGE Embedding Server (Port 8002)
+[program:bge]
+command=services/vllm/venv/bin/vllm serve --config services/vllm/config_bge.yaml
+directory=.
+autostart=false
+autorestart=true
+startretries=3
+stderr_logfile=.logs/bge.err.log
+stdout_logfile=.logs/bge.out.log
+stdout_logfile_maxbytes=50MB
+stdout_logfile_backups=10
+stderr_logfile_maxbytes=50MB
+stderr_logfile_backups=10
+environment=HF_HOME="./.cache/vllm",HF_TOKEN="%(ENV_HF_TOKEN)s"
+priority=210
+stopwaitsecs=30
+
+
 # AudioCraft Studio Service
 [program:audiocraft]
 command=services/audiocraft/venv/bin/python services/audiocraft/main.py
@@ -113,7 +131,7 @@ programs=comfyui,webdav-sync
 priority=100
 
 [group:vllm]
-programs=llama
+programs=llama,bge
 priority=200
 
 [group:audiocraft]