fix: empty-content guard + finish_reason=length handling + reasoning budget (v0.18.1)

Reasoning models at high effort could spend the entire max_tokens budget
on thinking and return zero visible content with finish_reason=length;
the empty article flowed writer -> editor (reject) -> empty draft post,
with the cost booked as success (2026-06-11 incident: prod run acf24029,
draft post 921; diagnosis: convo pipeline-v2-phase1 seq 12).

- Empty-completion guard at the provider seam (new
  OpenRouter_Completion_Guard): empty/whitespace content is a failure;
  one retry with reasoning disabled (per-call override) when
  finish_reason=length or reasoning was active, then the stage fails.
- finish_reason=length always warning-logged (model, stage, tokens).
- Failed/retried attempts booked to generation_log with
  response_status=error + real token burn; run_decisions verdict rows
  (retry_reasoning_disabled / failed_empty_completion) for v0.18.0 runs.
- Writer-path guard in Content_Generator::execute_stage and
  belt-and-braces empty check in Publisher::create_post.
- Reasoning budget sanity: reasoning.max_tokens cap + completion
  headroom via new prautoblogger_reasoning_max_tokens setting (default
  constant 2048, 0 = legacy uncapped effort mode); cap replaces effort
  when active; governor reservations use the effective ceiling.
- Body assembly moved to Request_Builder::build_body (300-line cap).
- Tests: guard verdict matrix, builder budget shaping, writer/publisher
  guards, full provider retry path.

Agent-Session: prab-empty-draft-guard-20260611

includes/admin/class-settings-fields-extended.php

@@ -78,7 +78,17 @@ class PRAutoBlogger_Settings_Fields_Extended {
 					'low'     => __( 'Low — light reasoning, lower cost', 'prautoblogger' ),
 					'minimal' => __( 'Minimal — barely any reasoning', 'prautoblogger' ),
 				),
-				'description' => __( 'How much effort the model spends reasoning before answering. Higher effort = more reasoning tokens = higher cost. Only applies when reasoning is enabled above.', 'prautoblogger' ),
+				'description' => __( 'How much effort the model spends reasoning before answering. Higher effort = more reasoning tokens = higher cost. Only applies when reasoning is enabled above. When the Reasoning Token Cap below is set (default), the cap is sent instead of the effort level — set the cap to 0 to use effort only.', 'prautoblogger' ),
+			),
+			array(
+				'id'          => 'prautoblogger_reasoning_max_tokens',
+				'label'       => __( 'Reasoning Token Cap', 'prautoblogger' ),
+				'type'        => 'number',
+				'section'     => 'prautoblogger_models',
+				'default'     => PRAUTOBLOGGER_DEFAULT_REASONING_MAX_TOKENS,
+				'min'         => 0,
+				'max'         => 32000,
+				'description' => __( 'Hard ceiling on thinking tokens per call when reasoning is enabled. The completion budget is raised by this amount so reasoning can never crowd out the article text — an uncapped high effort can spend the entire budget thinking and return an empty draft. 0 = no cap (uncapped effort mode, not recommended).', 'prautoblogger' ),
 			),
 
 			// ── Schedule & Budget ───────────────────────────────────────