refactor: fix critical bugs and improve project structure

Phase 1 - Critical Bug Fixes:
- Add tests/test_cuda_utils.h for GPU test skip support (blocking issue)
- Fix forward_with_timing() missing cudaStreamSynchronize()
- Unify launch_fused_gemm() error handling to throw std::invalid_argument
- Fix docs/releases/v1.1.0.md date from 2025 to 2024

Phase 1 - Cleanup:
- Remove .omc/ directory (was in .gitignore)
- Add .gitkeep to openspec/archive/ and openspec/changes/

Phase 2 - CI Optimization:
- Simplify CI format check from 4 action calls to single script
- Add test_autotuner.cpp, test_profiler.cpp, test_half_gemm.cu
- Add CODEOWNERS, ISSUE_TEMPLATE, SECURITY.md
- Update CMakeLists.txt with new test files

Phase 3 - AI Tooling:
- Add .clang-tidy configuration
- Enhance .clangd with ClangTidy support
- Enhance AGENTS.md with CUDA debugging guide and troubleshooting
- Enhance CLAUDE.md with error handling and test classification
- Simplify OpenSpec README files to pure index

New files: 12
Modified files: 7
Total changes: +692 -253 lines

Author: Release Bot

.clang-tidy

@@ -0,0 +1,29 @@
+---
+Checks: >
+  bugprone-*,
+  modernize-*,
+  performance-*,
+  readability-*,
+  -modernize-use-trailing-return-type,
+  -readability-magic-numbers,
+  -readability-identifier-length
+HeaderFilterRegex: '.*'
+CheckOptions:
+  - key: readability-identifier-naming.ClassCase
+    value: CamelCase
+  - key: readability-identifier-naming.FunctionCase
+    value: lower_case
+  - key: readability-identifier-naming.VariableCase
+    value: lower_case
+  - key: readability-identifier-naming.ConstantCase
+    value: UPPER_CASE
+  - key: readability-identifier-naming.ParameterCase
+    value: lower_case
+  - key: readability-identifier-naming.MemberCase
+    value: lower_case
+  - key: readability-identifier-naming.MemberSuffix
+    value: '_'
+  - key: performance-move-const-arg.CheckTriviallyCopyableMove
+    value: '0'
+  - key: bugprone-argument-comment.StrictMode
+    value: '1'

.clangd

@@ -10,6 +10,14 @@ CompileFlags:
 Diagnostics:
   UnusedIncludes: Strict
   MissingIncludes: Strict
+  ClangTidy:
+    Add:
+      - bugprone-*
+      - modernize-*
+      - performance-*
+      - readability-*
+    Remove:
+      - modernize-use-trailing-return-type
 
 Index:
   Background: Build

.github/CODEOWNERS

@@ -0,0 +1,87 @@
+# Code Owners
+
+This file defines the code ownership for the mini-inference-engine project.
+
+## Repository Owners
+
+* @shane
+
+## Code Ownership by Area
+
+### CUDA Kernels
+```
+src/*.cu
+include/kernels.cuh
+include/half_gemm.cuh
+include/vectorized_gemm.cuh
+```
+
+### Inference Engine
+```
+src/inference_engine.cpp
+include/inference_engine.h
+include/tensor.h
+src/tensor.cu
+```
+
+### Memory Management
+```
+include/memory_pool.h
+include/stream_manager.h
+```
+
+### Performance & Profiling
+```
+include/autotuner.h
+include/profiler.h
+benchmarks/
+```
+
+### Configuration & Logging
+```
+include/config.h
+include/logger.h
+```
+
+### Quantization
+```
+include/quantization.h
+```
+
+### Batch Operations
+```
+include/batch_gemm.h
+```
+
+### Tests
+```
+tests/
+```
+
+### Documentation
+```
+docs/
+README.md
+README.zh-CN.md
+CHANGELOG.md
+AGENTS.md
+CLAUDE.md
+```
+
+### CI/CD
+```
+.github/
+CMakeLists.txt
+CMakePresets.json
+```
+
+### OpenSpec
+```
+openspec/
+```
+
+## Review Requirements
+
+- Changes to core CUDA kernels require review by kernel owners
+- Breaking API changes require review from all area owners
+- Documentation changes can be approved by any owner

.github/ISSUE_TEMPLATE/bug_report.md

@@ -0,0 +1,55 @@
+---
+name: Bug Report
+about: Report a bug to help us improve
+title: '[BUG] '
+labels: bug
+assignees: ''
+---
+
+## Describe the Bug
+
+A clear and concise description of what the bug is.
+
+## Environment
+
+- OS: [e.g. Ubuntu 22.04]
+- CUDA Version: [e.g. 12.1]
+- GPU: [e.g. RTX 3080]
+- CMake Version: [e.g. 3.25]
+- Compiler: [e.g. GCC 11]
+
+## Steps to Reproduce
+
+1. Run command '...'
+2. With input '...'
+3. See error
+
+## Expected Behavior
+
+A clear and concise description of what you expected to happen.
+
+## Actual Behavior
+
+What actually happened.
+
+## Code Sample
+
+```cpp
+// Minimal reproducible code
+```
+
+## Error Output
+
+```
+Paste the error message or log output here
+```
+
+## Additional Context
+
+Add any other context about the problem here.
+
+## Checklist
+
+- [ ] I have searched existing issues to avoid duplicates
+- [ ] I have provided a minimal reproducible example
+- [ ] I have included all relevant environment information

.github/ISSUE_TEMPLATE/feature_request.md

@@ -0,0 +1,54 @@
+---
+name: Feature Request
+about: Suggest a new feature or enhancement
+title: '[FEATURE] '
+labels: enhancement
+assignees: ''
+---
+
+## Describe the Feature
+
+A clear and concise description of the feature you'd like to see.
+
+## Problem Statement
+
+What problem does this feature solve? Is this a new optimization technique, API improvement, or something else?
+
+## Proposed Solution
+
+How would you like this feature to work?
+
+### API Design (if applicable)
+
+```cpp
+// Proposed API signature
+void new_function(/* params */);
+```
+
+## Alternatives Considered
+
+Have you considered any alternative solutions or workarounds?
+
+## Use Case
+
+Describe a specific use case where this feature would be valuable.
+
+## Priority
+
+- [ ] Low: Nice to have
+- [ ] Medium: Would improve workflow
+- [ ] High: Blocking important use case
+
+## Implementation Notes
+
+Any technical considerations or constraints?
+
+## Additional Context
+
+Add any other context or screenshots about the feature request here.
+
+## Checklist
+
+- [ ] I have searched existing issues to avoid duplicates
+- [ ] I have provided a clear problem statement
+- [ ] I have considered alternative solutions

.github/workflows/ci.yml

@@ -85,29 +85,21 @@ jobs:
       - name: Checkout
         uses: actions/checkout@v4
 
-      - name: Check formatting (src)
-        uses: jidicula/clang-format-action@v4.18.0
-        with:
-          clang-format-version: '18'
-          check-path: 'src'
-
-      - name: Check formatting (include)
-        uses: jidicula/clang-format-action@v4.18.0
-        with:
-          clang-format-version: '18'
-          check-path: 'include'
-
-      - name: Check formatting (tests)
-        uses: jidicula/clang-format-action@v4.18.0
-        with:
-          clang-format-version: '18'
-          check-path: 'tests'
+      - name: Install clang-format
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y clang-format-18
 
-      - name: Check formatting (benchmarks)
-        uses: jidicula/clang-format-action@v4.18.0
-        with:
-          clang-format-version: '18'
-          check-path: 'benchmarks'
+      - name: Check formatting
+        run: |
+          echo "Checking code formatting..."
+          FILES=$(find src include tests benchmarks -type f \( -name '*.cpp' -o -name '*.cu' -o -name '*.h' -o -name '*.cuh' \) 2>/dev/null || true)
+          if [ -n "$FILES" ]; then
+            echo "$FILES" | xargs clang-format-18 --dry-run --Werror
+            echo "✓ All files formatted correctly"
+          else
+            echo "No source files found to check"
+          fi
 
   docs-check:
     name: Documentation Check