[TypeSpec Authoring] Support API example assessment

## Goal

Assess whether API examples remain complete and correct for the changed TypeSpec API surface.

## Scope

- Identify examples associated with newly added or materially changed operations and models.
- Detect missing examples for workflows where examples are expected.
- Detect stale examples that no longer match request or response schemas, parameters, required properties, enum values, defaults, errors, paging, or long-running behavior.
- Validate examples against generated artifacts or the TypeSpec semantic model where possible.
- Link findings to the relevant TypeSpec declaration and example file.
- Avoid duplicate findings when one shared example issue affects multiple operations.

## Acceptance criteria

- Missing examples for representative new workflows are detected.
- Stale request, response, parameter, enum, and model examples are detected.
- Valid examples do not produce findings.
- Findings clearly identify the required example update and include source links.
- Missing validation evidence produces an explicit blocked/not-assessed result.
- Tests cover missing, stale, valid, shared, and unavailable-evidence cases.

Start by mapping the TypeSpec declarations, example files, generated artifacts, and semantic model involved in API example assessment. Define how missing, stale, valid, shared, and unavailable-evidence cases should be represented, then add tests demonstrating the acceptance criteria and confirm findings include source links without duplicates.

Written by the indexing model from the issue text.