feat: consolidate most distros into "starter"

* Removes a bunch of distros
* Removed distros were added into the "starter" distribution
* Doc for "starter" has been added
* Partially reverts https://github.com/meta-llama/llama-stack/pull/2482
  since inference providers are disabled by default and can be turned on
  manually via env variable.
* Disables safety in starter distro

Closes: #2502
Signed-off-by: Sébastien Han <seb@redhat.com>

.github/workflows/integration-tests.yml

@@ -43,7 +43,7 @@ jobs:
 
       - name: Build Llama Stack
         run: |
-          uv run llama stack build --template ollama --image-type venv
+          uv run llama stack build --template starter --image-type venv
 
       - name: Check Storage and Memory Available Before Tests
         if: ${{ always() }}
@@ -54,16 +54,18 @@ jobs:
       - name: Run Integration Tests
         env:
           INFERENCE_MODEL: "meta-llama/Llama-3.2-3B-Instruct"
+          OLLAMA_INFERENCE_MODEL: "meta-llama/Llama-3.2-3B-Instruct" # for library tests
+          ENABLE_OLLAMA: "ollama" # for library tests
           OLLAMA_URL: "http://0.0.0.0:11434"
         run: |
           if [ "${{ matrix.client-type }}" == "library" ]; then
-            stack_config="ollama"
+            stack_config="starter"
           else
-            stack_config="server:ollama"
+            stack_config="server:starter"
           fi
           uv run pytest -s -v tests/integration/${{ matrix.test-type }} --stack-config=${stack_config} \
             -k "not(builtin_tool or safety_with_image or code_interpreter or test_rag)" \
-            --text-model="meta-llama/Llama-3.2-3B-Instruct" \
+            --text-model="ollama/meta-llama/Llama-3.2-3B-Instruct" \
             --embedding-model=all-MiniLM-L6-v2 \
             --color=yes \
             --capture=tee-sys | tee pytest-${{ matrix.test-type }}.log