fpm: calculate accurate process counts from states during status generation

Replace unreliable scoreboard counters with counts calculated from actual
process states. Since status endpoint already locks and reads all processes,
this adds zero performance overhead while ensuring 100% accurate metrics.

Resolves monitoring discrepancies where active processes were incorrectly
reported as significantly lower than actual non-idle processes under load.

Fixes #18956

Author: sylvesterdamgaard

sapi/fpm/fpm/fpm_status.c

@@ -65,15 +65,25 @@ int fpm_status_export_to_zval(zval *status)
 	struct fpm_scoreboard_proc_s procs[scoreboard.nprocs];
 
 	struct fpm_scoreboard_proc_s *proc_p;
+	int calculated_active = 0, calculated_idle = 0;
 	for(i=0; i<scoreboard.nprocs; i++) {
-		proc_p = fpm_scoreboard_proc_acquire(scoreboard_p, i, 1);
-		if (!proc_p){
-			procs[i].used=-1;
-			continue;
-		}
-		procs[i] = *proc_p;
-		fpm_scoreboard_proc_release(proc_p);
-	}
+            proc_p = fpm_scoreboard_proc_acquire(scoreboard_p, i, 1);
+            if (!proc_p){
+                procs[i].used=-1;
+                continue;
+            }
+            procs[i] = *proc_p;
+
+            if (proc_p->used) {
+                if (proc_p->request_stage == FPM_REQUEST_ACCEPTING) {
+                    calculated_idle++;
+                } else {
+                    calculated_active++;
+                }
+            }
+
+            fpm_scoreboard_proc_release(proc_p);
+        }
 	fpm_scoreboard_release(scoreboard_p);
 
 	now_epoch = time(NULL);
@@ -88,9 +98,9 @@ int fpm_status_export_to_zval(zval *status)
 	add_assoc_long(status, "listen-queue", scoreboard.lq);
 	add_assoc_long(status, "max-listen-queue", scoreboard.lq_max);
 	add_assoc_long(status, "listen-queue-len", scoreboard.lq_len);
-	add_assoc_long(status, "idle-processes", scoreboard.idle);
-	add_assoc_long(status, "active-processes", scoreboard.active);
-	add_assoc_long(status, "total-processes", scoreboard.idle + scoreboard.active);
+	add_assoc_long(status, "idle-processes", calculated_idle);
+	add_assoc_long(status, "active-processes", calculated_active);
+	add_assoc_long(status, "total-processes", calculated_idle + calculated_active);
 	add_assoc_long(status, "max-active-processes", scoreboard.active_max);
 	add_assoc_long(status, "max-children-reached", scoreboard.max_children_reached);
 	add_assoc_long(status, "slow-requests", scoreboard.slow_rq);