fix: Use SKU GPU Count in DeploymentSpec

[ishaansehgal99]

Reason for Change:
Cases were workspace was not utilizing all available GPUs on machine by default this was causing slower inference requests and unused resources. This is the fix.

[codecov]

Codecov Report

Attention: Patch coverage is 52.83019% with 25 lines in your changes missing coverage. Please review.

Project coverage is 57.00%. Comparing base (ed546f0) to head (a9273a1).
Report is 21 commits behind head on main.

Files	Patch %	Lines
pkg/utils/common.go	50.00%	17 Missing and 1 partial ⚠️
pkg/tuning/preset-tuning.go	37.50%	5 Missing ⚠️
pkg/inference/preset-inferences.go	77.77%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #541      +/-   ##
==========================================
- Coverage   61.10%   57.00%   -4.11%     
==========================================
  Files          29       30       +1     
  Lines        2602     2912     +310     
==========================================
+ Hits         1590     1660      +70     
- Misses        921     1157     +236     
- Partials       91       95       +4     

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