Bad upstream O2 sensor reading in 2005 Honda City iDSi 1.5 engine Indian versio

Hi,
I have recently purchased an old car, it is a Honda City 2005 idsi engine(dual sparks for each cylinder). Ever since I bought it, the fuel trims were negative only and were never positive, I cannot make a vacuum leak to force it to positive because it doesn't have a MAF sensor, instead it uses a MAP sensor. I have changed the downstream because it was completely dead when I purchased it and going through several videos I now suspect the O2 sensor or the circuit is bad. Till now I thought the bad was bad for O2 sensor, we even tried to connect chassis and block ground today and the trims were still the same. I check the bias voltage, it is around 400mv which seems normal and couldn't check the return to ecm signal voltage as there isnt a place to stick a wire on and also the o2 sensor is next to impossible to touch without a lift which I do not have. After connecting chassis to block ground, I want to ask what would happen if someone puts a downstream sensor in place of upstream sensor? I am not sure if the one that came with the car was OEM, the holes on the sensor are just like the downstream one. I also replace the sensor and same negative trims, I am using and recording trips with elm237 and no even once I have seen an positive trim. Considering that all aftermarket sensors downstream sold around here, is it possible that placing a downstream sensor in upstream make it read rich?
More info: With O2 in negative trim, it makes the car hesitate and fuel avg is half or even worse of than expected. When I disconnect and drive the car, the hesitation is gone with perfect pickup, this made me confirm this aint an engine issue and problem with O2 sensor reading it self. Please help.

How do the below trims look like?
O2 reads below .45mv too but still trims were -ve. is this normal?

I plugged the spare o2 I have with the sensor not plugged into CAT and it went positive, so O2 sensor and wiring is fine here.