fix: allow capacity http errors to requeue

[britaniar]

Description of your changes

When a querying the vm size recommender client, there may be times the client can run into transient client errors that will resolve on retry. At the moment the scheduler only runs once and sees all errors as unexpected. Now the possibility of transient http errors is possible with the addition of using azure property checker as we query an azure client. Therefore, these should not be considered unexpected, and the scheduler should be able to requeue on these errors.

Fixes #

I have:

• Updated the scheduler to requeue on http errors.

• Simplified http error logging to be one line for easier debugging.

• Run make reviewable to ensure this PR is ready for review.

How has this code been tested

Special notes for your reviewer

Simplified logging:

## BEFORE
E0612 02:18:53.167602 1 compute/vmsizerecommenderclient.go:122] "Failed to generate VM size re...
request failed with status 503: POST http://127.0.0.1:5859/compute/subscriptions/a5191316-d253-43e3...

## AFTER
E0623 12:38:21.725116  158733 vmsizerecommenderclient.go:121] "Failed to generate VM size recommendations" err="request failed with status 503: POST http://127.0.0.1:35911/subscriptions/sub-123/providers/Microsoft.Compute/locations/eastus/vmSizeRecommendations/vmAttributeBased/generate" subscriptionID="sub-123" location="eastus" clientRequestID="c68fbda3-fc22-400f-abd0-e15b4f7af56f" latencyMs=2