feat: Add approximate support for GeLU

[aturker-synnada]

Description

GeLU now takes approximate flag as init argument. Closes #35.

What is Changed

• GeLU approximate support added.

Checklist:

• Tests that cover the code added.
• Corresponding changes documented.
• All tests passed.
• The code linted and styled (pre-commit run --all-files has passed).

[mertbozkir]

Hello @aturker-synnada, if we add a return, could the CI build pass, right? Looks like fails because of return, anything that I should be aware of? 😅

The other ‘on_failure’ CI pipeline seems to require credentials. How should we handle that?

[aturker-synnada]

Hello @aturker-synnada, if we add a return, could the CI build pass, right? Looks like fails because of return, anything that I should be aware of? 😅

The other ‘on_failure’ CI pipeline seems to require credentials. How should we handle that?

Yes, you are right 😅. I guess it is not about credentials; it should do something when CI failed @mehmetozsoy-synnada. 

[aturker-synnada]

Fixed mypy error and updated GeLU init arguments.

[codecov]

Codecov Report

Attention: Patch coverage is 95.65217% with 3 lines in your changes missing coverage. Please review.

Project coverage is 86.51%. Comparing base (4ab5d44) to head (51f8f1c).
Report is 2 commits behind head on main.

Files with missing lines	Patch %	Lines
mithril/backends/with_autograd/mlx_backend/ops.py	25.00%	3 Missing ⚠️

@@            Coverage Diff             @@
##             main      #36      +/-   ##
==========================================
+ Coverage   86.48%   86.51%   +0.03%     
==========================================
  Files          61       61              
  Lines       15544    15600      +56     
==========================================
+ Hits        13443    13497      +54     
- Misses       2101     2103       +2     

Files with missing lines	Coverage Δ
mithril/backends/with_autograd/jax_backend/ops.py	84.81% <100.00%> (ø)
...ithril/backends/with_autograd/torch_backend/ops.py	79.26% <100.00%> (+0.05%)	⬆️
...hril/backends/with_manualgrad/numpy_backend/ops.py	85.13% <100.00%> (+0.12%)	⬆️
...backends/with_manualgrad/numpy_backend/ops_grad.py	86.73% <100.00%> (+0.06%)	⬆️
mithril/framework/constraints.py	94.80% <100.00%> (+0.12%)	⬆️
mithril/framework/logical/essential_primitives.py	99.57% <100.00%> (+<0.01%)	⬆️
mithril/models/primitives.py	97.42% <100.00%> (+<0.01%)	⬆️
mithril/backends/with_autograd/mlx_backend/ops.py	31.65% <25.00%> (-0.23%)	⬇️