[FEATURE] Add Bengali (Bangla) Support for Voice Input

### Preflight Checklist

- [x] I have searched [existing requests](https://github.com/anthropics/claude-code/issues?q=is%3Aissue%20label%3Aenhancement) and this feature hasn't been requested yet
- [x] This is a single feature request (not multiple features)

### Problem Statement

Claude Code supports Bengali in chat, but Bengali is not properly supported when using voice input.

When I enable voice input and speak in Bengali, the speech transcription does not accurately recognize or transcribe my Bengali speech. This makes it difficult to use voice input naturally when working in Bengali.

For example, if I say a Bengali instruction such as "এই ফাইলের মধ্যে একটা নতুন ফিচার যোগ করতে হবে", I would expect it to be transcribed as Bengali text, but the current voice input experience does not reliably support this.

As a Bengali-speaking developer and a Claude Code user on Mac, this is a significant limitation in my daily workflow. Bengali voice input already works well for me in ChatGPT, so without Bengali voice support in Claude Code, I may have to move my voice-based workflow to ChatGPT instead.

I would really appreciate it if Bengali voice input could be supported in Claude Code.

### Proposed Solution

Please add Bengali (Bangla / বাংলা) as a supported language for Claude Code's voice input and speech-to-text.

Ideally, when voice input is enabled, users should be able to speak naturally in Bengali and have their speech accurately transcribed into Bengali Unicode text, which can then be used directly as a Claude Code prompt.

The voice input system could allow Bengali to be selected alongside existing supported languages, or automatically detect Bengali speech when it is spoken.

For example:

Voice input: "এই ফাইলের মধ্যে একটা নতুন ফিচার যোগ করতে হবে"

Expected transcription:
"এই ফাইলের মধ্যে একটা নতুন ফিচার যোগ করতে হবে"

The goal is to make Bengali voice input work as naturally and reliably as English voice input.

### Alternative Solutions

_No response_

### Priority

Critical - Blocking my work

### Feature Category

CLI commands and flags

### Use Case Example

_No response_

### Additional Context

_No response_

The issue identifies voice input and speech-to-text in the Claude Code CLI but names no files, tests, or entry points. Start by locating the existing voice-input language-selection or transcription implementation, then determine how Bengali can be added. Done means Bengali speech is reliably transcribed as Bengali Unicode text through voice input.

Escrita pelo modelo de indexação a partir do texto da issue.