[Composite OPs] ConvTranspose2d

[mstojkovicTT]

Ticket

fixes: #2262

Problem description

explained in #2262

What's changed

Added support for nn.ConvTranspose2d.

Checklist

• Wait for MLIR PR to be merged and uplifted

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 66.56%. Comparing base (b53e143) to head (cabb0cf).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2629   +/-   ##
=======================================
  Coverage   66.56%   66.56%           
=======================================
  Files           5        5           
  Lines         326      326           
=======================================
  Hits          217      217           
  Misses        109      109           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[mstojkovicTT]

ConvTranspose2D, when wrapped to composite, gives the same results as just normally lowering it to TTIR from SHLO (trough convolution path with dilation checks). I dont think it makes sense to have this composite, at least for now.