Read secrets directly into GitHub secrets from ENV_VARS

[tectonia]

This pull request involves multiple changes focusing on the removal of .env file handling in GitHub Actions and updating the chat model deployment name. The most important changes include the removal of .env handling steps from various action files and workflows and moving to loading the secrets directly into environment variables, and updating the chat model deployment name from aoai to gpt-35-turbo.

Fixes #209

Removal of .env file handling:

• .github/actions/aml_real_deployment/action.yml: Removed the ENV_VARS input and associated steps for creating and loading .env files.
• .github/actions/kubernetes_deployment/action.yml: Removed steps for creating and loading .env files.
• .github/actions/webapp_deployment/action.yml: Removed the ENV_VARS input and steps for converting .env to environment variables.
• .github/workflows/platform_cd_dev_workflow.yml: Removed steps for creating and reading .env files and updated the method to load secrets directly from ENV_VARS. [1] [2] [3] [4]
• .github/workflows/platform_ci_dev_workflow.yml: Updated the method to load secrets directly from ENV_VARS instead of creating .env files.

Update chat model deployment name:

• chat_with_pdf/data/bert-paper-qna-1-line.jsonl: Changed CHAT_MODEL_DEPLOYMENT_NAME from aoai to gpt-35-turbo.
• chat_with_pdf/data/bert-paper-qna-3-line.jsonl: Changed CHAT_MODEL_DEPLOYMENT_NAME from aoai to gpt-35-turbo.
• chat_with_pdf/flows/evaluation/flow.dag.yaml: Updated deployment_name from aoai to gpt-35-turbo in multiple nodes. [1] [2] [3]
• chat_with_pdf/flows/standard/flow.dag.yaml: Changed CHAT_MODEL_DEPLOYMENT_NAME from aoai to gpt-35-turbo.