Reduce size of release binary

[andrews05]

This PR makes 3 changes that together reduce binary size by around 25%:

• Sets lto="fat" in cargo.toml
• Sets panic="abort" in cargo.toml
• Sets location-detail=none in RUSTFLAGS

Closes #571

An unrelated change: I've replaced the zopfli test file with a smaller one that runs much faster, as well as removing the slow test for issue-133 which was related to an older alpha optimisation that is no longer relevant.

[ace-dent]

@andrews05 - thanks for implementing this! Is the new image for testing permissively licensed?
Comments based on my recollection of #571 ...

1. You decided to avoid opt-level=s, which reduced execution speed ~10%. Now release will use the default 3 (it hasn't been explicitly stated). Seems reasonable.
2. In addition to lto=fat, I think it is useful to include codegen-units=1. This is missing?
3. I believe build-std with panic_immediate_abort were also found to be useful ?

Many thanks :-)

[andrews05]

The image is just a copy of another one in the test suite (see #129 - it comes from a corpus of png test images).

2. codegen-units=1 drastically slows down compile time for negligible benefit to file size, so it really doesn't seem worth it.
3. build-std is a much more advanced technique that I would definitely not recommend for a typical build process, unless there was a very specific need to reduce file size as much as possible. Since we don't have such a need, I'd much rather leave it up to end users to perform this if it's important to them.

[ace-dent]

Totally understand your points but it's a shame to not adopt (3), which seemed to save ~30% binary size.

[andrews05]

Yeah, I agree. It's a shame that rust can't avoid the bloat more easily.

[AlexTMjugador]

I think these changes look good, thanks!

Given that we're already using a nightly toolchain for the release builds, I don't think going with build-std is that difficult, but it surely comes with the tradeoff of slower CI times. We can explore that option in more depth in the future; the changes in this PR are not mutually exclusive with that.