Enable Overriding Full Container Image URLs with Environment Variables

[hdefazio]

Depends on #364

• Uses ${EPP_IMAGE} directly instead of ${IMAGE_TAG_BASE}:${EPP_TAG}
• Uses ${VLLM_SIMULATOR_IMAGE} directly instead of ${IMAGE_REGISTRY}/${VLLM_SIMULATOR_IMAGE}:${VLLM_SIMULATOR_TAG}
• Uses ${ROUTING_SIDECAR_IMAGE} directly instead of ghcr.io/llm-d/llm-d-routing-sidecar:${ROUTING_SIDECAR_TAG}
• Set VLLM_SIMULATOR_IMAGE and ROUTING_SIDECAR_IMAGE from the makefile for better visibility

Testing

$ make image-build 
✅ Container tool 'podman' found.
==== Building Docker image ghcr.io/llm-d/llm-d-inference-scheduler:dev ====
podman build \
....
Successfully tagged ghcr.io/llm-d/llm-d-inference-scheduler:dev
4287ce2db25f54dddb1add9066b32205c9e0184371be6ec9f0feda48b681b3f2

$ make image-pull 
✅ Container tool 'podman' found.
==== Pulling Docker images ====
./scripts/pull_images.sh
Using container tool: podman
--- Using the following images ---
Scheduler Image:     ghcr.io/llm-d/llm-d-inference-scheduler:dev
Simulator Image:     ghcr.io/llm-d/llm-d-inference-sim:latest
Sidecar Image:       ghcr.io/llm-d/llm-d-routing-sidecar:v0.2.0
----------------------------------------------------
Pulling dependencies...
Checking for image: ghcr.io/llm-d/llm-d-inference-scheduler:dev
 -> Image not found on registry. Checking for a local version...
 -> Found local-only image. Proceeding.
Checking for image: ghcr.io/llm-d/llm-d-inference-sim:latest
 -> Image found on registry. Pulling...
Trying to pull ghcr.io/llm-d/llm-d-inference-sim:latest...
Getting image source signatures
Copying blob f8829679f757 skipped: already exists  
Copying blob 1e02d32990ad skipped: already exists  
Copying blob 5a062cf5334b skipped: already exists  
Copying config 1409e399cd done   | 
Writing manifest to image destination
1409e399cd0a62efbe06de9927924e1bf7a68b3a9c3cee2d490f1eb504685b24
    ✅ Successfully pulled image.
Checking for image: ghcr.io/llm-d/llm-d-routing-sidecar:v0.2.0
 -> Image found on registry. Pulling...
Trying to pull ghcr.io/llm-d/llm-d-routing-sidecar:v0.2.0...
Getting image source signatures
Copying blob 24e43493af91 skipped: already exists  
Copying blob dd15713b0f22 skipped: already exists  
Copying config 2320a1dc7b done   | 
Writing manifest to image destination
2320a1dc7bbe987ddd07b5b7d9b7f14d0d6dcf9db9193b76372dcb81e3886faf
    ✅ Successfully pulled image.
Successfully pulled dependencies

$ make test-e2e
(this reruns the commands I already showed the output for above so I won't repeat it) 
==== Running End to End Tests ====
./test/scripts/run_e2e.sh
Running end to end tests
  "level"=0 "msg"="Successfully loaded environment variable" "key"="EPP_IMAGE" "value"="ghcr.io/llm-d/llm-d-inference-scheduler:dev"
  "level"=0 "msg"="Successfully loaded environment variable" "key"="VLLM_SIMULATOR_IMAGE" "value"="ghcr.io/llm-d/llm-d-inference-sim:latest"
  "level"=0 "msg"="Successfully loaded environment variable" "key"="ROUTING_SIDECAR_IMAGE" "value"="ghcr.io/llm-d/llm-d-routing-sidecar:v0.2.0"
  "level"=0 "msg"="Environment variable not set, using default value" "key"="EXISTS_TIMEOUT" "defaultValue"="30s"
  "level"=0 "msg"="Environment variable not set, using default value" "key"="READY_TIMEOUT" "defaultValue"="3m0s"
  "level"=0 "msg"="Environment variable not set, using default value" "key"="MODEL_READY_TIMEOUT" "defaultValue"="10m0s"
=== RUN   TestEndToEnd
Running Suite: End To End Test Suite - /home/hdefazio/projects/openshift-ai/llm-d-inference-scheduler/test/e2e
==============================================================================================================
Ran 3 of 3 Specs in 118.324 seconds
SUCCESS! -- 3 Passed | 0 Failed | 0 Pending | 0 Skipped
--- PASS: TestEndToEnd (118.32s)
PASS
ok      github.com/llm-d/llm-d-inference-scheduler/test/e2e     118.342s