Codechange: be more type-specific about types in NPFs queue

[TrueBrain]

Motivation / Problem

While working on #11190 the compiler (rightfully) complained a TileIndex was shoved in an uint32_t, and it didn't like that.

Description

Turns out, all key1 references are in fact always a TileIndex, and this is assumed in several places. So be more correct, and make it in fact a TileIndex. key2 is always Trackdir. Also rename key1 and key2, as .. weird names.

And just to be clear, this queue stuff in NPF is also used by rivers. Explaining the one change in landscape.cpp.

Limitations

Checklist for review

Some things are not automated, and forgotten often. This list is a reminder for the reviewers.

• The bug fix is important enough to be backported? (label: 'backport requested')
• This PR touches english.txt or translations? Check the guidelines
• This PR affects the save game format? (label 'savegame upgrade')
• This PR affects the GS/AI API? (label 'needs review: Script API')
  • ai_changelog.hpp, game_changelog.hpp need updating.
  • The compatibility wrappers (compat_*.nut) need updating.
• This PR affects the NewGRF API? (label 'needs review: NewGRF')
  • newgrf_debug_data.h may need updating.
  • PR must be added to API tracker

[PeterN]

Main question, rather than fiddling with variable types, is there a stl container that can be used instead of the custom queue?

[TrueBrain]

Most likely, but that would take far more time than I am willing to spend on this atm :) NPF is old .. Rivers is unknown to me .. testing takes a lot of time. I rather focus on getting the strong typedef in atm :)

[PeterN]

Might as well fix the type of key2 to be Trackdir as well then?