Fix for handling changes introduced in postprocess

[RomanSek]

Plugin postprocessing didn't introduce changes in files if string chain pointer didn't change. I removed additional check for that so HTTrack will now depend only on return value from plugin to apply changes.
Let me know if you need tests for this fix.

[xroche]

I don't get the change - if cAddr == TypedArrayElts(output_buffer), the copy will be idem-potent, so this does not change anything

[RomanSek]

I'm a bit rusty with my C, so I can be wrong. I suspect that bug happens when you realloc memory for changed output and get pointer to the same memory address, but with bigger/smaller size of the allocated memory block. In that case if cAddr == TypedArrayElts(output_buffer) will be true, but new output is different and isn't copied properly.

If you want - I can write a plugin that triggers this behavior.