fix(pioneer): add reasoning_options to reasoning models

Adds reasoning_options = [] to full-definition models with reasoning = true
(pioneer/auto, mistral-medium-3.5, Qwen3-8B, gpt-oss-120b, gpt-oss-20b).
Pioneer abstracts reasoning controls so no effort/toggle options are exposed.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

providers/pioneer/models/Qwen/Qwen3-8B.toml

@@ -4,6 +4,7 @@ release_date = "2025-03-31"
 last_updated = "2025-04-28"
 attachment = false
 reasoning = true
+reasoning_options = []
 temperature = true
 tool_call = true
 structured_output = true

providers/pioneer/models/mistral-medium-3.5.toml

@@ -4,6 +4,7 @@ release_date = "2026-03-31"
 last_updated = "2026-04-30"
 attachment = true
 reasoning = true
+reasoning_options = []
 temperature = true
 tool_call = true
 structured_output = true

providers/pioneer/models/openai/gpt-oss-120b.toml

@@ -4,6 +4,7 @@ release_date = "2025-07-31"
 last_updated = "2025-08-05"
 attachment = false
 reasoning = true
+reasoning_options = []
 temperature = true
 tool_call = true
 structured_output = true

providers/pioneer/models/openai/gpt-oss-20b.toml

@@ -4,6 +4,7 @@ release_date = "2025-07-31"
 last_updated = "2025-08-05"
 attachment = false
 reasoning = true
+reasoning_options = []
 temperature = true
 tool_call = true
 structured_output = true

providers/pioneer/models/pioneer/auto.toml

@@ -4,6 +4,7 @@ release_date = "2024-01-01"
 last_updated = "2025-06-29"
 attachment = true
 reasoning = true
+reasoning_options = []
 temperature = true
 tool_call = true
 structured_output = true