quota: z.ai credit-bucket back-pressure + resource gate + usage accounting (Redmine 490+491)

The loop now consults quota and host state before every dispatch and
DEFERS gated work with a logged reason instead of letting turns die at
the provider (the 2026-08-28 19:00 quota-wall failure mode, replayed as
a test). Adds internal/quota: 5h/weekly credit buckets (provider poll
when z.ai ships an endpoint - fake-server tested - else locally
estimated from the documented credit formula), TZ-aware peak window
(default 01:00-05:00 America/Chicago weekdays, matching the documented
z.ai peak Mon-Fri 14:00-18:00 Singapore), block/defer thresholds, a
read-only load/mem/disk/IO-PSI monitor, an optional redis shared-state
hop (stdlib RESP2 mini-client) so all instances of an account
coordinate, per-class token+credit accounting in loop.jsonl, and
`harness quota status|probe|gate`. Config: [quota] + [resources]
sections; README runbook covers the redis container and deploy-time
cgroup enforcement.

💘 Generated with Crush

Assisted-by: Crush:glm-5.2

internal/config/toml.go

@@ -10,6 +10,7 @@ import (
 	"fmt"
 	"os"
 	"sort"
+	"strconv"
 	"strings"
 )
 
@@ -48,6 +49,15 @@ func (d TOMLDoc) Bool(key string) (bool, bool) {
 	return v, ok
 }
 
+func (d TOMLDoc) Float(key string) (float64, bool) {
+	if v, ok := d[key].(float64); ok {
+		return v, true
+	}
+	// Bare integers read as floats too (max_load_avg = 6 parses as int64).
+	v, ok := d[key].(int64)
+	return float64(v), ok
+}
+
 func (d TOMLDoc) StringList(key string) ([]string, bool) {
 	raw, ok := d[key].([]any)
 	if !ok {
@@ -280,10 +290,13 @@ func parseScalar(v string, no int) (any, error) {
 		return false, nil
 	}
 	n, err := parseTOMLInt(v)
-	if err != nil {
-		return nil, fmt.Errorf("harness.toml:%d: unsupported value %q (want string, int, bool, or array)", no, v)
+	if err == nil {
+		return n, nil
 	}
-	return n, nil
+	if f, ferr := strconv.ParseFloat(strings.ReplaceAll(v, "_", ""), 64); ferr == nil {
+		return f, nil
+	}
+	return nil, fmt.Errorf("harness.toml:%d: unsupported value %q (want string, int, float, bool, or array)", no, v)
 }
 
 func parseTOMLInt(v string) (int64, error) {