Requesting either a --no-stream flag or a buffered output mode in 2.0. The old version printed faster. The new version is more visually disruptive

### Checks

- [x] I have searched [github.com/aws/amazon-q-developer-cli/issues](https://github.com/aws/amazon-q-developer-cli/issues?q=) and there are no duplicates of my issue
- [x] I have run `q doctor` in the affected terminal session
- [x] I have run `q restart` and replicated the issue again

### Operating system

Linux 6.6.87.2-microsoft-standard-WSL2 - Ubuntu 22.04.5 LTS

### Expected behaviour

Chat responses should either render instantly (buffered until complete) or provide a configuration option to control streaming output. The previous kiro-cli 1.x rendered responses noticeably faster and was easier to read.

### Actual behaviour

In kiro-cli 2.0.0, chat responses render token-by-token as they stream from the model. The output scrolls too slowly to feel instant but too fast to actually read along with, resulting in a poor reading experience. There is no config command, --no-stream flag, or any setting to control this behaviour. --classic seems to help a little bit but then you lose the new TUI, which looks nice and creates more readable breaks and divisions (when not actively printing)

### Steps to reproduce

1. Install kiro-cli 2.0.0
2. Run kiro-cli chat
3. Ask any question that produces a multi-paragraph response
4. Observe the token-by-token rendering — text scrolls continuously at an awkward pace that is difficult to follow
5. Run kiro-cli chat --help — no streaming-related options exist

### Environment

```yaml

[q-details]
version = "2.0.0"
hash = "0026b0505ef7bc2b10780c7bdb57813e5d949747"
date = "2026-04-13T09:03:52.798806Z (1d ago)"
variant = "minimal"

[system-info]
chip = "13th Gen Intel(R) Core(TM) i7-1355U"
total-cores = 6
memory = "15.44 GB"

[system-info.os.linux]
kernel_version = "6.6.87.2-microsoft-standard-WSL2"
id = "ubuntu"
name = "Ubuntu"
pretty_name = "Ubuntu 22.04.5 LTS"
version_id = "22.04"
version = "22.04.5 LTS (Jammy Jellyfish)"

[environment]
cwd = "/home/USER"
cli-path = "/home/USER"
os = "Linux"
shell-path = "/usr/bin/bash"
shell-version = "5.1.16"
install-method = "unknown"
in-wsl = true

[env-vars]
DISPLAY = ":0"
PATH = "/home/USER/.nvm/versions/node/v24.14.0/bin:/home/USER/bin:/home/USER/.nvm/versions/node/v24.14.0/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/usr/lib/wsl/lib:/mnt/c/Python313/Scripts/:/mnt/c/Python313/:/mnt/c/WINDOWS/system32:/mnt/c/WINDOWS:/mnt/c/WINDOWS/System32/Wbem:/mnt/c/WINDOWS/System32/WindowsPowerShell/v1.0/:/mnt/c/WINDOWS/System32/OpenSSH/:/mnt/c/Program Files/PuTTY/:/mnt/c/Program Files/nodejs/:/mnt/c/ProgramData/chocolatey/bin:/mnt/c/Program Files/Amazon/AWSCLIV2/:/mnt/c/Program Files/Docker/Docker/resources/bin:/mnt/c/Users/connellybr/AppData/Local/VanDyke Software/Clients/:/mnt/c/Users/connellybr/AppData/Local/VanDyke Software/SecureFX/:/mnt/c/Users/connellybr/AppData/Local/Microsoft/WindowsApps:/mnt/c/Users/connellybr/AppData/Local/Programs/Git/cmd:/mnt/c/Users/connellybr/AppData/Local/Programs/Git/mingw64/bin:/mnt/c/Users/connellybr/AppData/Local/Programs/Git/usr/bin:/mnt/c/Users/connellybr/AppData/Local/Programs/Microsoft VS Code/bin:/mnt/c/bin:/mnt/c/Users/connellybr/AppData/Roaming/npm:/mnt/c/Users/connellybr/AppData/Local/GitHubDesktop/bin:/mnt/c/Users/connellybr/AppData/Local/Programs/Kiro/bin:/snap/bin:/home/USER/.local/bin:/home/USER/.local/bin:/home/USER/bin"
QTERM_SESSION_ID = "caf891fa643f4be0a5fea1d53fc5440e"
Q_SET_PARENT_CHECK = "1"
Q_TERM = "1.26.1"
SHELL = "/bin/bash"
TERM = "xterm-256color"
```

Beginne beim Befehl `kiro-cli chat` und vergleiche seinen Streaming-Pfad mit dem Verhalten von `--classic`; verwende `kiro-cli chat --help`, um nachzuverfolgen, wo die Chat-Optionen definiert sind. Als abgeschlossen gilt die Bereitstellung einer dokumentierten Möglichkeit, gepufferte oder nicht gestreamte Ausgabe auszuwählen und dabei die neue TUI beizubehalten, wobei Antworten mit mehreren Absätzen in einem gut lesbaren Tempo dargestellt werden.

Vom Indexierungsmodell aus dem Issue-Text verfasst.