[CLI] Model just repeats the same line over and over in chat

## Problem
I installed phi-4-reasoning-generic-gpu
This prompt: List the main scientific achievements of the last 1,000 years
Produced this output (I finally did ctrl+c):

User asks: "List the main scientific achievements of the last 1,000 years". We need to produce a list of main scientific achievements of the last 1,000 years. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements. We need to produce a list of achievements

I ran this from the CLI with: _foundry chat phi-4-reasoning-generic-gpu_

## Diagnostics

## System
- **OS:** Microsoft Windows 10.0.26200
- **Architecture:** X64
- **CPU:** Intel(R) Core(TM) Ultra 7 165U (14 logical cores)
- **GPU:** Intel Intel(R) Graphics (128.0 MB), Intel Intel(R) Graphics (128.0 MB)
- **NPU:** Intel Corporation Intel(R) AI Boost
- **RAM:** 27.2 GB available / 63.6 GB total
- **Disk:** 448.0 GB free / 951.6 GB total (C:\)
- **.NET:** .NET 9.0.18

## Service
- **State:** Ready
- **Web URLs:** http://127.0.0.1:57998
- **CLI version:** 0.10.3
- **Foundry Local Core:** 1.0.0
- **ORT:** 1.26.0
- **ORT GenAI:** 0.0.0

## Models
- **Available variants:** 111
- **Loaded:** 1
- **Cached:** 1

## Connectivity
- **Local service:** Reachable
- **Model registry:** Reachable

## Warnings
- No warnings.

Reproduce the issue with `foundry chat phi-4-reasoning-generic-gpu` using the prompt and Windows diagnostics from the report. Trace the CLI chat and local inference entry points to determine where repeated output originates; done means the model produces a normal response instead of repeating the same line indefinitely.

Written by the indexing model from the issue text.