feat(v2): add ReasoningLevel option; thinking/reasoning across providers

Introduces an opt-in level-based reasoning toggle (low/medium/high) that
each provider translates to its native parameter:

- Anthropic: thinking.budget_tokens (1024/8000/24000), with temperature
  forced to default and MaxTokens auto-grown above the budget.
- OpenAI/xAI/Groq via openaicompat: reasoning_effort string, gated by a
  new Rules.SupportsReasoning predicate so non-reasoning models don't
  receive the parameter. xAI uses Rules.MapReasoningEffort to remap
  "medium" to "high" since its API only accepts low|high.
- Google: thinking_config.thinking_budget + include_thoughts:true.
- DeepSeek: SupportsReasoning=false (reasoner is always-on; the
  reasoning_content trace was already extracted via openaicompat).

Reasoning content is surfaced as Response.Thinking on Complete and as
StreamEventThinking deltas during streaming. Provider-side: extracted
from Anthropic thinking content blocks, Google's part.Thought=true
parts, and the non-standard reasoning_content field that DeepSeek and
Groq emit (parsed out of raw JSON since openai-go doesn't type it).

Public API:
  - llm.ReasoningLevel + ReasoningLow/Medium/High constants
  - llm.WithReasoning(level) request option
  - Model.WithReasoning(level) for baked-in defaults
  - provider.Request.Reasoning, provider.Response.Thinking
  - provider.StreamEventThinking

Tests cover Rules-based gating, MapReasoningEffort, reasoning_content
extraction (Complete + Stream), Anthropic budget mapping, and
temperature suppression when thinking is enabled. Existing behavior is
unchanged when Reasoning is the empty string.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

v2/provider/provider.go

@@ -80,6 +80,15 @@ type Request struct {
 	// CacheHints requests prompt-cache breakpoints at specified positions
 	// on providers that support it (currently Anthropic). nil = no caching.
 	CacheHints *CacheHints
+
+	// Reasoning, when non-empty, asks the model to spend extra inference
+	// budget reasoning before answering. Each provider translates this to
+	// its native parameter (Anthropic thinking.budget_tokens, OpenAI/xAI
+	// reasoning_effort, Google thinking_config, etc.). Models that do not
+	// support reasoning silently ignore it.
+	//
+	// Allowed values: "" (no reasoning, default), "low", "medium", "high".
+	Reasoning string
 }
 
 // Response is a completion response at the provider level.
@@ -87,6 +96,11 @@ type Response struct {
 	Text      string
 	ToolCalls []ToolCall
 	Usage     *Usage
+
+	// Thinking holds the model's reasoning/thinking trace, when one was
+	// requested and the provider exposed it. Empty for providers/models
+	// that do not surface a thinking trace.
+	Thinking string
 }
 
 // Usage captures token consumption.
@@ -117,6 +131,7 @@ const (
 	StreamEventToolEnd                          // Tool call complete
 	StreamEventDone                             // Stream complete
 	StreamEventError                            // Error occurred
+	StreamEventThinking                         // Reasoning/thinking content delta
 )
 
 // StreamEvent represents a single event in a streaming response.