fix(mistral-ai): forward response_format for JSON schema

[arun2728]

Description: (required)

• Register response_format on the Mistral AI chat completion ProviderConfig so the gateway forwards OpenAI-compatible response_format (including type: "json_schema" and nested json_schema) to Mistral’s /v1/chat/completions API instead of dropping it during transformUsingProviderConfig.
• When response_format.type is json_schema and the inner json_schema object omits name, set a default name of "response" so payloads match Mistral’s OpenAPI (required name + schema on the inner object) while staying compatible with clients that omit name.

Fixes #1559

Tests Run/Test cases added: (required)

• Ran transformUsingProviderConfig(MistralAIChatCompleteConfig, …) with response_format.type: "json_schema" and verified the transformed body includes response_format and defaulted json_schema.name when omitted.
• End-to-end: curl to local gateway (x-portkey-provider: mistral-ai) with response_format / JSON schema and confirmed a successful structured response from Mistral.

Type of Change:

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• Refactoring (no functional changes)

[arun2728]

Hi @roh26it, just following up on the PR I raised last week. The fix is non-breaking and only addresses the issue without affecting existing functionality. Please take a look whenever you get a chance. Thanks!