## Summary
Relate to #17284. Completes the batch 5 migration of 7 OpenAI-compatible
drivers (`vllm`, `volcengine`, `xai`, `xiaomi`, `xinference`, `xunfei`,
`zhipu-ai`) onto the unified request/response helpers
(`doRequest`/`doStreamRequest` +
`HandleNonStreamingResponse`/`HandleStreamingResponse` +
`ParserConfig`), established by `deepseek` in #17634.
This branch is rebased on the current `pr/migrate-models-batch5` and
fixes the issues in the previous state of the PR.
Co-authored-by: Haruko386 <tryeverypossible@163.com>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
### What problem does this PR solve?
Adds a shared safe default implementation for unsupported Go
model-driver capability methods and migrates the confirmed panic-stub
providers to use it.
The Go `ModelDriver` interface requires providers to implement many
capability methods even when the provider does not support them. XunFei
had unsupported capability methods implemented as `panic("implement
me")`, Mistral still had a panic in `ParseFile`, and HuaweiCloud carried
an unreachable `panic("implement me")` after a normal chat return.
### Type of change
- [x] Refactoring
Co-authored-by: Haruko386 <tryeverypossible@163.com>
