Refactor eating morale modifiers and messages

[wapcaplet]

Summary

SUMMARY: Infrastructure "Move food morale modifiers to modify_morale"

Purpose of change

Purely code refactoring/cleanup - separation of responsibility, and improved readability.

Describe the solution

It moves sections of player::eat concerned only with modifying morale to the more appropriate function, Character::modify_morale, and refactors a couple sections.

Describe alternatives you've considered

Considered leaving the morale effects in player::eat.

Testing

Used vim to ensure no undefined variables were left around.

Regression testing: Used debugging menu in game to mutate / spawn items and eat various morale-modifying foods, and ensure the correct messages and morale effects happened, including but not limited to:

• Eat at a table, with/without table manners
• Eat junk food with a sweet tooth (positive morale) or as a carnivore (negative morale)
• Eat human flesh as a cannibal (+psycho, +spiritual)
• Eat honey as an Ursine

Additional context

This was intended as part of a patch for #7533 but that's a separate problem (and harder for me to solve), so I thought it worthwhile on the basis of cleanup alone.

[codemime]

This is a very good change. However, some further improvements can be made to the moved code.

[wapcaplet]

This is a very good change. However, some further improvements can be made to the moved code.

@codemime Thanks for the suggestions! I will clean these up the best I can.

The part about looking for a nearby table is almost duplicated (along with the multiple invocations of veh_at) in two other source files as well, with the only difference being those are looking within a radius of 2 instead of 1:

Cataclysm-DDA/src/activity_handlers.cpp

Lines 412 to 419 in 6017c7c

	bool has_table_nearby = false;
	for( const tripoint &pt : g->m.points_in_radius( u.pos(), 2 ) ) {
	if( g->m.has_flag_furn( flag_FLAT_SURF, pt ) || g->m.has_flag( flag_FLAT_SURF, pt ) ||
	( ( g->m.veh_at( pt ) && ( g->m.veh_at( pt )->vehicle().has_part( flag_KITCHEN ) ||
	g->m.veh_at( pt )->vehicle().has_part( flag_FLAT_SURF ) ) ) ) ) {
	has_table_nearby = true;
	}
	}

Cataclysm-DDA/src/activity_item_handling.cpp

Lines 1428 to 1435 in 6017c7c

	bool has_table_nearby = false;
	for( const tripoint &pt : g->m.points_in_radius( src_loc, 2 ) ) {
	if( g->m.has_flag_furn( flag_FLAT_SURF, pt ) || g->m.has_flag( flag_FLAT_SURF, pt ) ||
	( ( g->m.veh_at( pt ) && ( g->m.veh_at( pt )->vehicle().has_part( flag_KITCHEN ) ||
	g->m.veh_at( pt )->vehicle().has_part( flag_FLAT_SURF ) ) ) ) ) {
	has_table_nearby = true;
	}
	}

So it's clear we may need a standalone has_table_nearby function, with a distance argument - should that be a separate PR, or would it be appropriate to roll it into this one? I would prefer not to touch more files than necessary, so my plan for now is to tidy up the morale code that I moved, and consider opening a new PR to refactor those other uses of has_table_nearby.

[wapcaplet]

New has_nearby_table function factored out and merged in PR #37790 - today I will finish cleaning up and testing modify_morale using the new function.

[wapcaplet]

I've done another pass of regression testing and this looks ready to go.

There's a travis-ci failure but I don't think I can do anything about it:

Reading package lists...
Err:4 http://ppa.launchpad.net/ubuntu-toolchain-r/test/ubuntu xenial InRelease
  Could not connect to ppa.launchpad.net:80 (91.189.95.83), connection timed out
Err:5 http://apt.llvm.org/xenial llvm-toolchain-xenial-8 InRelease
  Could not connect to apt.llvm.org:80 (151.101.250.49), connection timed out
W: Failed to fetch http://apt.llvm.org/xenial/dists/llvm-toolchain-xenial-8/InRelease  Could not connect to apt.llvm.org:80 (151.101.250.49), connection timed out
W: Failed to fetch http://ppa.launchpad.net/ubuntu-toolchain-r/test/ubuntu/dists/xenial/InRelease  Could not connect to ppa.launchpad.net:80 (91.189.95.83), connection timed out
W: Some index files failed to download. They have been ignored, or old ones used instead.