Fix flaky tests for azsdk-cli by removing non-deterministic behavior

[Copilot]

This PR fixes two flaky tests in the azsdk-cli test suite that were causing inconsistent failures in the CI pipeline:

Issues Fixed

1. TestSimpleFailCase - Random Exit Code Generation

The test was using new Random().Next(2) to generate a random expected exit code (0 or 1), but the actual implementation always returns exit code 1. This created a race condition where the test would:

• Pass when random generated 1 (matching actual behavior)
• Fail when random generated 0 (not matching actual behavior)

Before:

var expectedExitCode = new Random().Next(2);
var expected = $"[ERROR] RESPONDING TO 'HI. MY NAME IS' with FAIL: {expectedExitCode}";

After:

var expected = "[ERROR] RESPONDING TO 'HI. MY NAME IS' with FAIL: 1";

2. TestWindows - Incorrect Expected Output

The test expected "Result: nul" but the actual output was "Result: null", causing consistent failures on Windows platforms.

Before:

var expected = @"
Message: RESPONDING TO 'HI. MY NAME IS' with SUCCESS: 0
Result: nul
Duration: 1ms".TrimStart();

After:

var expected = @"
Message: RESPONDING TO 'HI. MY NAME IS' with SUCCESS: 0
Result: null
Duration: 1ms".TrimStart();

Testing

• Verified tests now pass consistently across multiple runs
• Confirmed all existing tests in the suite continue to pass
• Changes are minimal and surgical, maintaining the original test intent

The fixes eliminate non-deterministic behavior while preserving the core functionality being tested.

Fixes #46.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.