cont

This release includes improvements to ggml, specifically shortening virtual device naming in CUDA and Metal. It also updates the device description build process for ggml-metal and refines naming conventions.

This release focuses on internal improvements to the ggml graph and sampler logic, including stricter handling of view outputs and bug fixes for sampler and continuous inference.

This release stabilizes the interaction between the lazy grammar sampler and the active reasoning budget by inhibiting sampling when reasoning is active. It also includes various bug fixes and provides updated binaries for numerous hardware and OS configurations.

This release focuses on internal fixes within the graph and continuous modules, specifically addressing issues related to KQ mask reuse and adapter checks.

This release enhances ggml with extended binary broadcast support for permuted source 1 and stabilizes continuous tensor handling by ensuring s0 is always 1.

This release updates the server component to utilize different seeds for child completions and includes handling for the default seed within the continuation logic.