Raise default agent token budget for reasoning models

The tracker mapping, common-pattern enrichment, and third-party
disambiguation agents default to a small max-tokens budget on the
premise that their final output is tiny structured JSON. On
reasoning models such as the GPT-5 family, reasoning tokens count
against max_tokens, so a small budget is consumed by reasoning and
the JSON is truncated, surfacing as "unexpected end of JSON input".

Raise the defaults to 4096 (1024 -> 4096 for tracker mapping, 512
-> 4096 for disambiguation) to leave headroom for the reasoning
phase. Update the bootstrap builder default, its test, and the
production values example to match.

Signed-off-by: Émile Ré <emile@probo.com>

pkg/cookiebanner/tracker_mapping_agent.go

@@ -40,10 +40,14 @@ const (
 	defaultMappingMaxTurns    = 4
 	defaultEnrichmentMaxTurns = 3
 
-	// defaultAgentMaxTokens caps the structured output of the mapping
-	// and enrichment agents when the agent config carries no max-tokens
-	// budget. Both outputs are tiny structured JSON.
-	defaultAgentMaxTokens = 1024
+	// defaultAgentMaxTokens caps the output of the mapping and
+	// enrichment agents when the agent config carries no max-tokens
+	// budget. Both final outputs are tiny structured JSON, but the
+	// budget must leave ample headroom for reasoning models (e.g. the
+	// GPT-5 family): their reasoning tokens count against max_tokens,
+	// so too small a budget gets consumed by reasoning and truncates
+	// the JSON, surfacing as "unexpected end of JSON input".
+	defaultAgentMaxTokens = 4096
 
 	agentThirdPartyConfidenceThreshold = 0.6
 	// agentSourceConfidence is the fixed confidence stored on catalog