Currently, when running terraform plan in Harness, the -lock=true flag is enforced in the command, and attempts to override it via TF_CLI_ARGS_plan environment variable do not take effect.

We attempted to override the -lock flag by setting TF_CLI_ARGS_plan=-lock=false both as an environment variable and as a stage variable. However, the argument shown in the command in step logs remains -lock=true, and the state still gets locked during the plan step.

This will allow teams to control locking behavior during plan (e.g., disable locking for read-only plan runs while keeping it enabled for apply runs).

Without this capability, plan steps in Harness unnecessarily lock the state, potentially blocking other operations.