- Fix visibility errors I stupidly didn't test before committing earlier
  today
- Silence warnings about casting that were necessary for ARM
- Update dependencies

Update Cargo dependencies

Travis didn't like the old versions.

It wasn't the Arc unwrap causing the crash on Linux. Maybe it's this.

Something about these seemed to be causing a crash on Travis (build 327)... I have no idea what would set it off, but this makes the code better anyway.

- Users v0.5.1, which renames OSUsers to UsersCache
- Locale v0.2, which returns to libc v0.1
- Datetime v0.4.2, which mimics the locale update, and puts timezone definitions in:
- Zoneinfo-data, which is needed to obtain the current timezone

The `--long` flag should show the '@' character in the permissions list if that feature has been compiled in, but only the `--extended` flag should actually show their keys, rather than just their presence.

This makes use of a change in the `users` crate to change which parts of exa's code are accessed under a `Mutex`. The change is that the methods on `Users` can now take just `&self`, instead of `&mut self`. This has a knock-on effect in exa, as many methods now don't need to take a mutable `&self`, meaning that the Mutex can be moved to only containing the users information instead of having to be queried for *every column*. This means that threading should now be a lot faster, as fewer parts have to be executed on a single thread.

The main change to facilitate this is that `Table`'s structure has changed: everything environmental that gets loaded at the beginning is now in an `Environment` struct, which can be mocked out if necessary, as one of `Table`'s fields. (They were kind of in a variety of places before.)

Casualties include having to make some of the test code more verbose, as it explicitly takes the columns and environment as references rather than values, and those both need to be put on the stack beforehand. Also, all the colours are now hidden behind an `opts` field, so a lot of the rendering code is more verbose too (but not greatly so).

Document using `cargo install` in README

Cargo now lets people install binaries by running `cargo install`. This
can be more convenient than cloning and building exa manually.

This commit separates the code used to generate the tree structure characters from the code used to build tables, meaning that it'll become possible to display tree structures without using any of the table code.

Also, some tests are added to make sure that the tree code *basically* works.

1. imports from std
2. imports from external crates
3. imports from local modules
4. imports from self

This commit moves the colours module to be a sub-module of the output one.
This makes sense because finding which colour a certain file should be is only
done during output, and (I think) the only places that the `Colours` struct's
fields are ever queried is from the output module.

The only casualty was that the `file_colour` from the filetype module had to
be moved, as determining colours is no longer part of that module - only
determining filetype is. So it now reflects its name!

The benefit of this is that it make it possible to convert text cell contents
vectors into text cells with a method (see next commit). Casualties include
having to call `.into()` on vectors everywhere, which I'm not convinced is a
bad thing.

Because, strictly speaking, it's not a length, it's a width!

Also, re-order some struct constructors so that they're no longer
order-dependent (it's no longer the case that a value will be borrowed for one
field then consumed in another, meaning they have to be ordered in a certain
way to compile. Now the value is just worked out beforehand and the fields can
be specified in any order)

By removing the `File#file_name_width` method, we can make the file module
have no dependency on the output module -- in other words, the model (file)
and the view (output) are now separate again!

This commit introduces the `output::cell::DisplayWidth` struct, which
encapsulates the Unicode *display width* of a string in a struct that makes it
less easily confused with the *length* of a string.

The use of this type means that it's now harder to accidentally use a string's
length-in-bytes as its width. I've fixed at least one case in the code where
this was being done!

The only casualty is that it introduces a dependency on the output module from
the file module, which will be removed next commit.

A recent change to ansi-term [1] means that `ANSIString`s can now hold either
owned *or* borrowed data (Rust calls this the Cow type). This means that we
can delay formatting ANSIStrings into ANSI-control-code-formatted strings
until it's absolutely necessary. The process for doing this was:

1. Replace the `Cell` type with a `TextCell` type that holds a vector of
   `ANSIString` values instead of a formatted string. It still does the
   width tracking.

2. Rework the details module's `render` functions to emit values of this
   type.

3. Similarly, rework the functions that produce cells containing filenames
   to use a `File` value's `name` field, which is an owned `String` that
   can now be re-used.

4. Update the printing, formatting, and width-calculating code in the
   details and grid-details views to produce a table by adding vectors
   together instead of adding strings together, delaying the formatting as
   long as it can.

This results in fewer allocations (as fewer `String` values are produced), and
makes the API tidier (as fewer `String` values are being passed around without
having their contents specified).

This also paves the way to Windows support, or at least support for
non-ANSI terminals: by delaying the time until strings are formatted,
it'll now be easier to change *how* they are formatted.

Casualties include:

- Bump to ansi_term v0.7.1, which impls `PartialEq` and `Debug` on
  `ANSIString`.
- The grid_details and lines views now need to take a vector of files, rather
  than a borrowed slice, so the filename cells produced now own the filename
  strings that get taken from files.
- Fixed the signature of `File#link_target` to specify that the
  file produced refers to the same directory, rather than some phantom
  directory with the same lifetime as the file. (This was wrong from the
  start, but it broke nothing until now)

References:

[1]: ansi-term@f6a6579ba8174de1cae64d181ec04af32ba2a4f0

Fix compilation on arm using stable rust 1.5

Minor fixes for FreeBSD

src/feature/xattr.rs:6:5: 6:22 warning: unused import, #[warn(unused_imports)] on by default
src/feature/xattr.rs:6 use std::ffi::CString;
                           ^~~~~~~~~~~~~~~~~

$ exa --help
[...]
  -@, --extended     display extended attribute keys and sizes

$ exa -@
Unrecognized option: '@'.

$ exa --extended
Unrecognized option: 'extended'.

$ exa --git
Option --git is useless without option --long.

$ exa -l --git
.rw-r--r--  11k user 10 Dec 18:26 -- Cargo.lock
[...]

src/term.rs:37:39: 37:49 error: unresolved name `TIOCGWINSZ` [E0425]
src/term.rs:37     let result = ioctl(STDOUT_FILENO, TIOCGWINSZ, &mut window);
                                                     ^~~~~~~~~~

GitHub's new, wider view makes two screenshots side-to-side look nicer than one.