Added automated tool testing framework (#126)

* first version for system test workflow

* re-organized output and transient directories

* added pre/post testing scripts support 

* added sepatare modules for RAG, EFS, EVS

* updated contribution guidelines, logging...

* split rag test cases from core, added option to yield error in `create_response` utility, simplified output.

Author: remi-td

.gitignore

@@ -11,3 +11,4 @@ env
 test_results/
 custom*_objects.yaml
 CLAUDE.md
+var/

docs/CONTRIBUTING.md

@@ -2,7 +2,10 @@
 Make sure you have setup your environment based on the Developer Guide in this repo. The goal is to allow contributions to this project by anyone, and that all code requirements are automated. Here are the guidelines we adhere to as a team.
 
 ## Development Guidelines
-- Always ask before creating a PR. To avoid duplication efforts, its better to discuss it with us first or create an issue.
+- Always engage on the discussion board or create an issue before creating a PR. 
+- All PRs must have at least one issue associated.
+- Run testing before PR, and copy/paste the test report status in the PR. You can simply run the mandatory test tools with `python tests/run_mcp_tests.py "uv run teradata-mcp-server"`. For more information see [our testing guide](/tests/README.md)
+- Create a new test case if you add a new tool. For more information see [our testing guide](/tests/README.md)
 - All code must be reviewed via a pull request. Before anything can be merged, it must be reviewed by at least 2 others. [Contributing to a project step by step instuctions](https://docs.github.com/en/get-started/exploring-projects-on-github/contributing-to-a-project)
 - Squash commits into a single commit for your PR. We want to keep a clean git history.
 - Code should adhere to lint and codestyle tests. While you can commit code that doesn't validate but still works, it is encouraged to validate your code. It saves other's headaches down the road.