(experiment) llm tools: position before user prompt and after

[mialso]

Description:

• LLM: experiment tools different positions: before user prompt and after

tool call correctness test (no cache) ✔️ :

bare examples/toolCalling.js

Results: pass ✔️

• 'Qwen3-1.7B-Q4_0.gguf' model - the results are the same, regardless tools go before or after the user prompt

dynamic tool call cache test ❌ :

cd ./packages/qvac-sdk
bun run build:pack
bun run examples/llamacpp-dynamic-tools.ts

Results: fail ❌

• 'Qwen3-1.7B-Q4_0.gguf' model - dynamic tool calls invalidates the cache, model doesn't remember it's previous answer

[mialso]

implemented #706 and #745

[github-actions]

Tier-based Approval Status

**PR Tier:** TIER1

**Current Status:** ❌ PENDING

**Requirements:**
- 1 Team Member approval ❌ (0/1)
- 1 Team Lead OR Management approval ❌ (0/1)



---
*This comment is automatically updated when reviews change.*