Bug: Tortoise stuck in BackToNormal status for VPA-only deployments after emergency mode change

[AVSBharadwaj]

Description

A bug has been identified in Tortoise's resource management when handling deployments that use only vertical scaling policies (without horizontal scaling).

Issue Details

Tortoise gets stuck in BackToNormal status and fails to transition to Working phase
Resource recommendations are not being applied correctly
Results in over-provisioning of CPU resources and inflated resource allocations
Affects multiple services using VPA-only configurations

Technical Context

This issue appeared after the recent "emergency mode" change introduced in v1.1.0
The new logic does not properly handle scenarios without Horizontal Pod Autoscaler (HPA)
Error logs show "invalid container resource metric" messages

Workaround

Deleting and recreating the Tortoise resource temporarily resolves the issue, or reverting to v0.11.0.

Proposed Solution

We need to update the code to properly handle the transition from BackToNormal to Working state when only vertical scaling policies are configured.