"Can someone explain to me why ECU didn't report this event as Ignition Coil failure, instead of Injector failure?"
Let me play the Devil's Advocate here.
The ECU Reports an Injector Failure.
It is found that the Plug Caps have a problem.
The Plug Caps are repaired.
Under certain power settings. the Fuel Injector fault returns.
Is this a case of Confirmation Bias?
The Plug Caps had a problem that might have been causing RF Electrical Noise, but Not any ignition failure.
Nothing to detect.
An injector draws a current that is out of the normally expected value and is reported as a fault.
Because a Plug Cap was found to be faulty, This Confirms that the Injector was not the problem. (???)
Is it possible that there are two unrelated issues at play?
The Plug Caps were burned but caused no detectable problem (yet!)
The Fuel Injector has a problem that has not been identified.
The ECU can only report what it can detect at the far end of the circuit.
The ECU intermittently sees an improper current in the injector circuit and reports it.
Note the term "Injector Circuit".
It could be the actual Injector, a connector, the wiring, a fuse connection, or anything else in the circuit.
All of these items are in series and appear as a single fault to the ECU.
If the ECU is reporting an Injector Fault, it would be false logic to assume the Plug Cap is the real cause.
Re: Confirmation Bias