Add multi-arch install test workflow

[Joeavaikath]

🏗️ Add Cross-Architecture Build & Test Workflow

📋 Overview

This PR introduces a comprehensive cross-architecture CI/CD pipeline that builds and tests the OADP CLI across 6 different architectures using GitHub's native runners. This ensures our kubectl plugin works correctly on all major platforms where users deploy it.

🎯 What This Adds

🔧 Multi-Architecture Build & Test Pipeline

• Builds once, tests everywhere: Single build job creates binaries for all platforms
• Native execution testing: Each binary runs on its target hardware architecture
• Real kubectl plugin validation: Tests actual installation and plugin functionality
• Comprehensive error handling: Explicit error checking with clear failure messages

🌍 Platform Coverage

Platform	Architecture	Runner	Status
Linux	amd64	ubuntu-latest	✅ Native testing
Linux	arm64	ubuntu-24.04-arm	✅ Native testing
macOS	amd64	macos-13 (Intel)	✅ Native testing
macOS	arm64	macos-latest (Apple Silicon)	✅ Native testing
Windows	amd64	windows-latest	✅ Native testing
Windows	arm64	windows-11-arm	✅ Native testing

🚀 Key Features

Build Phase

• Uses existing make release-build to build all 6 architectures
• Validates all binaries are created successfully
• Uploads platform-specific artifacts for parallel testing

Test Phase

• Native execution: Each binary runs on its target platform's hardware
• kubectl plugin testing: Validates kubectl oadp functionality
• Help system validation: Tests all command help outputs
• Installation testing: Verifies plugin installs correctly

Quality Assurance

• Explicit error handling: set -e for bash, $ErrorActionPreference = "Stop" for PowerShell
• Fail-fast behavior: Immediate exit with clear error messages on any failure
• Comprehensive logging: Each step reports success/failure with descriptive output

🔧 Technical Improvements

Fixed Windows Testing

• ✅ Windows .exe extension: Test binaries now properly handle .exe suffix
• ✅ Native ARM64 testing: Uses windows-11-arm runner for true ARM64 validation
• ✅ PowerShell error handling: Try-catch blocks with proper exception reporting

Enhanced Test Infrastructure

• ✅ Cross-platform binary naming: Handles platform-specific executable extensions
• ✅ Robust error detection: File existence checks before attempting operations
• ✅ Clear failure reporting: Descriptive error messages for debugging

GitHub Actions Optimization

• ✅ Efficient artifact strategy: Separate artifacts per platform for parallel downloads
• ✅ Matrix optimization: Platform-specific configurations for optimal testing
• ✅ Build status integration: Added workflow badge to README

📊 Workflow Structure

graph TD
    A[Build All Architectures] --> B[Linux Testing]
    A --> C[macOS Testing] 
    A --> D[Windows Testing]
    B --> E[Test Summary]
    C --> E
    D --> E

Loading

Jobs:

1. build-all: Creates binaries for all 6 architectures using make
2. test-linux: Native testing on Ubuntu x64 & ARM64 runners
3. test-macos: Native testing on Intel & Apple Silicon runners
4. test-windows: Native testing on Windows x64 & ARM64 runners
5. test-summary: Aggregates results and reports overall status

🎉 Benefits

For Users

• ✅ Confidence: All released binaries are validated on target hardware
• ✅ Compatibility: Catches architecture-specific issues before release
• ✅ Quality: Every platform gets the same thorough testing

For Developers

• ✅ Early detection: CI catches cross-platform issues immediately
• ✅ Debugging: Clear error messages pinpoint exactly what failed where
• ✅ Automation: No manual testing needed across different architectures

For Releases

• ✅ Release confidence: Every binary is validated before distribution
• ✅ Platform parity: Ensures consistent experience across all platforms
• ✅ Krew readiness: Validates kubectl plugin functionality for krew distribution

🧪 Testing

This workflow validates:

• ✅ Cross-compilation: All 6 binaries build successfully
• ✅ Native execution: Each binary runs correctly on target hardware
• ✅ Help system: All command help outputs work properly
• ✅ Plugin installation: kubectl plugin functionality across platforms
• ✅ Test suite: Full Go test suite passes on each platform

📈 Files Changed

• +442 lines: New comprehensive cross-architecture workflow
• tests/common.go: Windows .exe extension support
• .gitignore: Refined pattern for generated manifests
• README.md: Added workflow status badge

🏁 Result

This PR establishes gold-standard cross-architecture CI that ensures the OADP CLI works flawlessly across all major platforms where Kubernetes runs, providing confidence for both developers and users.

---

Ready for production-quality releases! 🚀

[kaovilai]

While GitHub Actions runs with set -e by default (which causes the script to exit on the first non-zero exit code), it would be better to make the error handling explicit here.

Currently, if any of these kubectl-oadp commands fail, the success message will not be printed due to the default behavior, but this isn't immediately obvious from reading the code.

Consider using one of these approaches for clarity:

Option 1 - Chain with &&:

echo "Testing kubectl plugin functionality on ${{ matrix.arch }}..." && \
kubectl-oadp --help && \
kubectl-oadp version --help && \
kubectl-oadp nonadmin --help && \
echo "✅ kubectl plugin tests passed on ${{ matrix.arch }}"

Option 2 - Use if statement:

echo "Testing kubectl plugin functionality on ${{ matrix.arch }}..."
if kubectl-oadp --help && \
   kubectl-oadp version --help && \
   kubectl-oadp nonadmin --help; then
  echo "✅ kubectl plugin tests passed on ${{ matrix.arch }}"
else
  echo "❌ kubectl plugin tests failed on ${{ matrix.arch }}"
  exit 1
fi

This makes the error handling explicit and ensures the success message only appears when all commands succeed.

[Joeavaikath]

Explicit error handling added, ty!

[kaovilai]

Same error handling concern as above. Consider making the error handling explicit with && chaining or an if statement.

[kaovilai]

For future enhancement, you can https://github.com/docker/setup-qemu-action or podman equivalent and perform similar tests for ppc640le or s390x