[Feature Request] Create tools trust configurations for each profile

### 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

macOS 15.3.2 (24D81)

### Expected behaviour

The ability to configure tool permissions at the profile level, similar to having both "global tools trust profile" and "profile level tools trust profile". This would allow granular control over which tools can be executed based on the active profile.

### Actual behaviour

Currently, the same set of MCP tools are configured globally for all profiles without profile-level control over trust configurations.

```
[chat-transcript]
> /help

> /issue

I'd be happy to help you report an issue or make a feature request. You can use the `report_issue` function to submit your feedback directly.

To help make your report as effective as possible, could you please provide:

1. A title for your issue or feature request
2. What behavior you're experiencing (for issues) or what feature you'd like to request
3. Any steps to reproduce the issue (if applicable)
4. What you expected to happen (if different from what actually happened)

Once you provide these details, I can help you submit the report.
[Tool uses: none]

> Title of the issue: Create tools trust configurations for each profile

I understand you'd like to report an issue with the title "Create tools trust configurations for each profile". I'll help you submit this feature request. Could you please provide some additional details about:

1. What specific behavior or functionality you're requesting
2. How you expect this feature to work
3. Any current limitations you're experiencing that this feature would address

This information will help make your feature request more actionable for the development team.
[Tool uses: none]

> 2. I'd like to be able to configure tool permissions at the profile level. I hav econfigured the same set of mcp tools for all profiles globally but want profile-level control over which profile has trust configurations for certain tools. This will allow me to have granular control over which tools to execute based on my profile. Ex. I want to be able to assume admin credentials using a tool (just an example:)) for beta environments but not for gamma/prod. I want to be able to configure this at my profile level. 2. Similar to a "global context" and "profile level context". I'd like a "global tools trust profile", then a "profile level tools trust profile" 3. I don't want to grant default trust permissions for high risk tools - it increases some churn during tool execution.

I'll help you submit this feature request with the detailed information you've provided. Let me use the report_issue function to create a GitHub issue for this feature request.
[Tool uses: report_issue]
```

### Steps to reproduce

User has configured MCP tools globally but wants to restrict certain high-risk tools to specific profiles (e.g., allow admin credential tools for beta environments but not for gamma/prod).

### Environment

```yaml

[q-details]
version = "1.8.0"
hash = "dd7442517f2f878fde45c744863412868eee7574"
date = "2025-04-23T18:52:24.776083Z (5d ago)"
variant = "full"

[system-info]
os = "macOS 15.3.2 (24D81)"
chip = "Intel(R) Core(TM) i7-9750H CPU @ 2.60GHz"
total-cores = 6
memory = "16.00 GB"

[environment]
cwd = "/Users/USER"
cli-path = "/Users/USER"
os = "Mac"
shell-path = "/bin/zsh"
shell-version = "5.9"
terminal = "iTerm 2"
install-method = "unknown"

[chat-settings]
interactive=true

[chat-trusted_tools]

[chat-failed_request_ids]
none

[chat-context]
current_profile=default
profiles=
default

global_context=
.amazonq/rules/**/*.md
README.md
AmazonQ.md

profile_context=none

files=none
```

Inizia tracciando lo stato del profilo mostrato nel report: current_profile, profiles, global_context, profile_context e chat-trusted_tools. L’issue non indica file, punti di ingresso o test; l’implementazione completata dovrebbe fornire configurazioni di attendibilità separate a livello globale e di profilo, in modo che gli strumenti ad alto rischio possano essere abilitati per un profilo senza concedere attendibilità agli altri.

Scritto dal modello di indicizzazione a partire dal testo della issue.