ci : refactor (#23789)

* ci : separate CUDA windows workflow + fix names

* ci : rename workflow

* ci : prefix cache names with workflow name

* ci : rename build.yml -> build-cpu.yml

* ci : cache keys

* ci : fix windows cuda/hip concurrency of release workflow

* ci : fix apple cache names

* ci : add TODOs

* cont : keep just the last cache

* ci : update release concurrency to queue

* ci : move the release trigger to ubuntu-slim

* ci : hip add TODO

* cont : improve words

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

.github/workflows/build-webgpu.yml

@@ -35,7 +35,7 @@ env:
   LLAMA_ARG_LOG_TIMESTAMPS: 1
 
 jobs:
-  macos-latest-webgpu:
+  macos:
     runs-on: macos-latest
 
     steps:
@@ -46,7 +46,7 @@ jobs:
       - name: ccache
         uses: ggml-org/ccache-action@v1.2.21
         with:
-          key: macos-latest-webgpu
+          key: webgpu-macos-latest
           evict-old-files: 1d
           save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
 
@@ -76,7 +76,7 @@ jobs:
           cd build
           ctest -L main --verbose --timeout 900
 
-  ubuntu-24-webgpu:
+  ubuntu:
     runs-on: ubuntu-24.04
 
     steps:
@@ -87,7 +87,7 @@ jobs:
       - name: ccache
         uses: ggml-org/ccache-action@v1.2.21
         with:
-          key: ubuntu-24-webgpu
+          key: webgpu-ubuntu-24.04
           evict-old-files: 1d
           save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
 
@@ -129,8 +129,16 @@ jobs:
           # test-backend-ops is too slow on llvmpipe, skip it
           ctest -L main -E test-backend-ops --verbose --timeout 900
 
-  ubuntu-24-webgpu-wasm:
-    runs-on: ${{ 'ubuntu-24.04-arm' || 'ubuntu-24.04' }}
+  ubuntu-wasm:
+    strategy:
+      matrix:
+        include:
+          - build: 'x64'
+            os: ubuntu-24.04
+          - build: 'arm64'
+            os: ubuntu-24.04-arm
+
+    runs-on: ${{ matrix.os }}
 
     steps:
       - name: Clone
@@ -140,7 +148,7 @@ jobs:
       - name: ccache
         uses: ggml-org/ccache-action@v1.2.21
         with:
-          key: ubuntu-24-webgpu-wasm
+          key: webgpu-${{ matrix.os }}-wasm
           evict-old-files: 1d
           save: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}