chore: add overflow tests for unsigned types in BigCast (#5866)

Signed-off-by: cancaicai <2356672992@qq.com>

fix: Move from macos-13 runners to macos-15-intel runners (#5875)

Signed-off-by: Robert Kruszewski <github@robertk.io>

fix: Correctly pin maturin to 1.10.2 (#5873)

Signed-off-by: Robert Kruszewski <github@robertk.io>

fix: enable all features for vortex-mask docsrs and enable serde-deri…

…ve when serde is enabled (#5871)

Signed-off-by: Robert Kruszewski <github@robertk.io>

Add advisory semver checks to CI (#5645)

Also allow for custom pre-release drafts.

---------

Signed-off-by: Nicholas Gates <nick@nickgates.com>

chore: All of vortex-gpu requires compiler flag (#5362)

Signed-off-by: Robert Kruszewski <github@robertk.io>

Pipeline execution (#5251)

Signed-off-by: Nicholas Gates <nick@nickgates.com>

FIx java bindings packaging (#5008)

Seems to work
[here](https://github.com/vortex-data/vortex/actions/runs/18659973729/job/53197925411).
My general thinking is that `setup-protoc` runs on the provided
`container` config that's required only for the JNI build. That image
lacks unzip, so we just install it.

Signed-off-by: Adam Gutglick <adam@spiraldb.com>

Refresh some flatbuffer code (#4745)

Regenerated with a new `flatc` version which generates nicer more
2024-edition-friendly edition code.

---------

Signed-off-by: Adam Gutglick <adam@spiraldb.com>

Dispatch each I/O request (#4464)

An attempt at short-term fix for
#4400

The coalescing driver was spawned as a single task, and so landed on a
single thread of the I/O dispatcher regardless of how many threads you
gave it. This PR spawns each read request to allow other threads a
chance to play ball.

The reason the Polars patch doesn't work is that we need to continue to
dispatch work in order to provide a Tokio runtime within non-Tokio
clients, such as DuckDB. This is a bigger change as part of #4406