Fix queue panel always showing 0 and jobs API crash

Hey, For a long time, the queue panel in the UI has been completely broken on this fork: - Queue count always stays at 0 - Running/queued jobs never appear in the list - Browser console shows repeated 500 errors on `/get_jobs` I was still able execute normally my workflows and queue more than once, I just didn't had indication of how many were queued and what jobs are currently queued. After a lot of time trying to figure this out and searching the logs, I finally found the problem. **What happened:** This fork switched the internal queue items from old 5 element tuples to the new dict-based format (with `priority`, `id`, `extra_data`, etc..) probably when the per-queue preview stuff landed. But `comfy_execution/jobs.py` was never updated. Both `normalize_queue_item()` and `normalize_history_item()` were still trying to unpack tuples with `item[0]`, `item[1]`, etc., which fails hard on dicts (KeyError: 0). **The fix:** Made `normalize_queue_item()` handle both the old tuple format and the new dict format your fork uses. Also cleaned up the history one for consistency. One small change, zero side effects. Tested on Windows with the exact startup command I use daily - queue counter now updates and shows running jobs again. When I first had this issue, I checked in the issues section and I didn't see anyone report this, so I figure it might be a problem in my machine, but I tried reinstalling it multiple times, so I'm doing something wrong or nobody cared enough to report it. If you don't have this problem, and this is just a fix for my machine please tell me. And yes, this is a problem with the nightly version and the latest release. It's been a problem in my end for several versions.