#14862: fp32 support in unary

[KalaivaniMCW]

Ticket

Link to Github Issue #14862

Problem description

Provide context for the problem.

What's changed

Enabled preserve_fp32_precision flag for float32 input

Checklist

• Post commit CI passes https://github.com/tenstorrent/tt-metal/actions/runs/11752673780
  https://github.com/tenstorrent/tt-metal/actions/runs/11797734586
• Nightly fd https://github.com/tenstorrent/tt-metal/actions/runs/11797739127
• Model perf
• Device perf
• Demo tests
• Blackhole Post commit (if applicable)
• Model regression CI testing passes (if applicable)
• Device performance regression CI testing passes (if applicable)
• New/Existing tests provide coverage for changes

[eyonland]

Let's make sure that we don't loose any performance. But looks good.

[dmakoviichuk-tt]

Thanks!
Please add a few more random tests for a couple of ops.

[rfurko-tt]

Thanks for enabling FP32 support for unary ops. Looks good overall.
Please create separate test for each unary op to make sure that this fix works.