[arm-common-types]: Add a model for string encoding

### Service Team Contacts (Dev)

jovinson

### Choose which kind of change this is

Add common-type

### Describe the change, Provide the desired TypeSpec or json-schema for the change as appropriate.

There's [newer guidance](https://github.com/microsoft/api-guidelines/blob/vNext/azure/Guidelines.md#returning-string-offsets--lengths-substrings) for data plane APIs on returning substring offsets or lengths within a text. Our service has [implemented a type](https://github.com/Azure/azure-rest-api-specs/blob/9016a56fdc5c9dc140796c1acc38e4373755c500/specification/healthdataaiservices/HealthDataAIServices.DeidServices/models_tagger.tsp#L31) that encapsulates the three encoding-specific values, but this seems like a good candidate for a common type since any other team handling text offsets will need to recreate the same type.

### Link to documentation if available

_No response_

### Additional notes or questions

_No response_

### Checklist

- [x] Follow our [Code of Conduct](https://github.com/azure/typespec-azure/blob/main/CODE_OF_CONDUCT.md)
- [x] Change is approved by Arm API Review
- [x] Please update issue title and choose the change type.

Start with the linked specification/healthdataaiservices/HealthDataAIServices.DeidServices/models_tagger.tsp model to understand the three encoding-specific values, then inspect the repository's common-type conventions. Done means adding a reusable common model that represents those values for text offsets and confirming it matches the stated API guidance.

Written by the indexing model from the issue text.