upgrading to v2

[yb175]

currently pullshark is able to generate test cases but lacks in the system failure analysis , the version 2 would focus on :

• what part of systems can affect ?
• how cost and latency can be improved ?
• what failure mode can happen in real production ?
• test senerios related to system reliability

so it would act as senior engineer pointing out system failures

## 🦈 PullShark AI Risk Analysis

**Risk Level**: 🔴 High  
*(Based on system area touched and historical bug patterns)*

### 🌐 System Impact
This PR modifies background job retry behavior, which affects how failed tasks are reprocessed in production.

### 🧠 Why This PR Looks Risky
- Retry logic changed without visible idempotency safeguard
- Touches code path previously linked to race-condition issues

### 💥 Potential Failure Scenarios
- Duplicate job execution under network retries
- Partial state updates if retry occurs mid-processing

### 👀 Reviewer Focus
- Confirm idempotency keys or deduplication logic exist
- Check behavior when downstream service times out

### 🧪 Tests That Would Catch Production Failures
- [ ] Simulate job retry after partial failure
- [ ] Concurrent retry scenario under load

### 🐛 Similar Issues Seen Before
- Payment processing race condition → Add transaction locks and idempotency keys

---
*PullShark highlights system & failure risks — not style issues.*