Add GitHub Pull Request template

Co-authored-by: torosent <[email protected]>

Author: Copilot

.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,128 @@
+<!--
+`durabletask-dotnet` Pull Request Template
+
+Fill in all sections. If a section does not apply, write "N/A".
+PRs that change runtime behavior must include manual validation steps and results.
+-->
+
+# Summary
+
+## What changed?
+<!-- 1 to 5 sentences. What does this PR do? -->
+
+## Why is this change needed?
+<!-- Link to bug/feature request and describe the problem being solved. -->
+
+## Related issues / work items
+- Fixes #
+- Related #
+
+---
+
+# Type of change
+- [ ] Bug fix
+- [ ] New feature
+- [ ] Performance improvement
+- [ ] Reliability / resiliency improvement
+- [ ] Refactor (no behavior change intended)
+- [ ] Test-only change
+- [ ] Build / CI change
+- [ ] Documentation-only change
+
+---
+
+# AI-assisted code disclosure (required)
+
+## Was an AI tool used? (select one)
+- [ ] No, this PR was written without AI assistance
+- [ ] Yes, AI helped write parts of this PR (e.g., GitHub Copilot)
+- [ ] Yes, an AI agent generated most of this PR
+
+## If AI was used, complete the following
+- Tool(s) used:
+- Which files / areas were AI-assisted:
+- What you changed after AI generation (review, refactor, bug fixes):
+
+### AI verification checklist (required if AI was used)
+- [ ] I understand the code in this PR and can explain it
+- [ ] I verified all referenced APIs/types exist and are correct
+- [ ] I reviewed edge cases and failure paths (timeouts, retries, cancellation, exceptions)
+- [ ] I reviewed concurrency/async behavior (no deadlocks, no blocking waits, correct cancellation tokens)
+- [ ] I checked for unintended breaking changes or behavior changes
+
+---
+
+# Testing
+
+## Automated tests
+### What did you run?
+<!-- Examples: `dotnet test`, specific test projects, filters -->
+- Command(s):
+
+### Results
+- [ ] Passed
+- [ ] Failed (explain and link logs)
+
+### Tests added/updated in this PR
+<!-- Briefly list new or updated tests, or write N/A -->
+- 
+
+---
+
+## Manual validation (required for runtime/behavior changes)
+> If this is docs-only or test-only, explain why manual validation is N/A.
+
+### Environment
+- OS:
+- .NET SDK/runtime version:
+- DurableTask component(s) affected (client/worker/backend/etc.):
+
+### Scenarios executed (check all that apply)
+- [ ] Orchestration start, completion, and replay behavior
+- [ ] Activity execution (including retries)
+- [ ] Failure handling (exceptions, poison messages, transient failures)
+- [ ] Cancellation and termination flows
+- [ ] Timers and long-running orchestration behavior
+- [ ] Concurrency / scale behavior (multiple instances, parallel activities)
+- [ ] Backward compatibility check (old history / upgraded worker) if applicable
+- [ ] Other (describe):
+
+### Steps and observed results (required)
+<!-- Provide exact steps so a reviewer can reproduce.
+Include relevant logs/trace snippets or describe what you observed. -->
+1.
+2.
+3.
+
+Evidence (logs, screenshots, traces, links):
+- 
+
+---
+
+# Compatibility / Breaking changes
+
+- [ ] No breaking changes
+- [ ] Breaking changes (describe below)
+
+If breaking:
+- Impacted APIs/behavior:
+- Migration guidance:
+- Versioning considerations:
+
+---
+
+# Review checklist (author)
+
+- [ ] Code builds locally
+- [ ] No unnecessary refactors or unrelated formatting changes
+- [ ] Public API changes are justified and documented (XML docs / README / samples as appropriate)
+- [ ] Logging is useful and not noisy (no secrets, no PII)
+- [ ] Error handling follows existing DurableTask patterns
+- [ ] Performance impact considered (hot paths, allocations, I/O)
+- [ ] Security considerations reviewed (input validation, secrets, injection, SSRF, etc.)
+
+---
+
+# Notes for reviewers
+<!-- Anything that helps reviewers: design notes, tradeoffs, follow-ups, risky areas -->
+-