P124 - Shadow integer issue slot

P124 adds a one-entry shadow integer issue slot. The first version accidentally kept the copied P123 smoke-script markers, so the shell completed but the harness waited for the wrong P124-FILE-OK marker. After retagging the scripts, I also tightened the model so it decodes the queued fetchq_ir_q instruction instead of the older in-flight instruction.

The corrected BusyBox shell workload passed:

P124-FILE-OK @ cycle 218,328,284
post-load cycles  : 218,328,141
shell window      : 64,450,052
instr retired     : 86,212,975
CPI               : 2.5324

The issue-slot model found:

candidates          : 40,458,763
accepts             : 9,802,160
issues              : 9,802,160
hold cycles         : 11,295,134
full blocks         : 3,289,938
dependency blocks   : 11,304,142
memory blocks       : 6,658,522
control blocks      : 9,327,439
system blocks       : 76,562
unknown blocks      : 0

That makes the next backend step more specific: source-ready tracking and class-specific holding records matter before a larger ROB or real scheduler will mean anything.