#15266: add sweeps based on traces

[bbradelTT]

Ticket

Link to Github Issue #15266

Problem description

We need a way to track progress towards generality for mlir / pytorch2.0

What's changed

Add sweeps based on traces for reduce, fused, and matmul ops. Running these and seeing the results will help with tracking.

Tested locally that all the tests run and failures are expected.

Checklist

• [x] Post commit CI passes. Doesn't touch anything in post commit. Between https://github.com/tenstorrent/tt-metal/actions/runs/11942061016 and https://github.com/tenstorrent/tt-metal/actions/runs/11942440734 everything passed.
• [ ] Blackhole Post commit (if applicable) N/A
• [ ] Model regression CI testing passes (if applicable) N/A
• [ ] Device performance regression CI testing passes (if applicable) N/A
• [x] New/Existing tests provide coverage for changes