CPU operator optimization

The goal of this issue is to highlight the need to improve kernels execution speed as backend CPU is used a lot for protoyping. Slow kernels make this task a bit painful.

• Conv[DepthWise]
• Arithmetic operators

changed milestone to %aidge_backend_cpu v0.4.0

added EnhancementPerformance ⚡️ TopicOperator labels

marked this issue as related to #5

mentioned in merge request !101 (merged)

added StatusReview Ready label

marked the checklist item Conv[DepthWise] as completed

added StatusDelayed label and removed StatusReview Ready label