Restore regression to green after commit 1's RTL strip. Drop dead-signal
references from 10 TBs to match the post-strip RTL surface:
- chirp TBs: drop clk_100m / reset_100m_n / new_elevation / new_azimuth
port wiring on plfm_chirp_controller_v2; drop elevation/azimuth counter
wires + check() assertions; retire Group 9 (Beam steering counters) in
tb_chirp_controller and C3 (stm32 toggle non-drive) in tb_chirp_contract.
- system TBs: drop stm32_new_chirp / stm32_new_elevation / stm32_new_azimuth
regs + DUT port wiring; drop current_elevation / current_azimuth wires;
delete stm32_chirp_toggle helper task; retire G6.1 (radar_mode opcode
0x01), G6.6 (trigger_pulse opcode 0x02), G7.1 (STM32->FPGA chirp CDC
stress), and G14.1/.2/.3 (range_mode opcode 0x20) test groups; replace
stm32_chirp_toggle calls in G2 with passive waits (scheduler auto-scans).
- receiver TB: drop .host_mode / .host_range_mode / .host_trigger ties.
- usb TBs: drop status_radar_mode / status_range_mode regs + DUT port
wiring + apply_reset init + per-test setup; update word[4][1:0] expected
value from 2'b10 (range_mode=2) to 2'b00 (now reserved) in
tb_usb_data_interface; refresh T3.7 comment in tb_usb_protocol_v2
(assertion value 0x60 unchanged because range_mode was already 2'd0).
- adc pwdn TB: strip host_radar_mode dummy reg + opcode 0x01 dispatch from
the local dispatch_block mirror; rewrite T5 with unknown opcode 0x05 to
preserve the dispatch-isolation case after 0x01 retirement.
Regression: 42/0/0 (was 37/0/6 in the pre-strip baseline where scipy was
unavailable; with radar_venv on PATH all 42 phases run and pass).
0da5ac6eaa