- Passing `ILogger` into executors

# Conflicts:
#	LLama.Web/Common/ModelOptions.cs
#	LLama.Web/Services/ConnectionSessionService.cs
#	LLama/LLamaStatelessExecutor.cs
#	LLama/LLamaWeights.cs

Major llama.cpp API Change

- Fixed stateless executor out-of-context handling
 - Fixed token tests

- Added ifdef to remove `Deconstruct` extension on everything except `NETSTANDARD2_0`

- Covered text transforms
 - Removed unnecessary non-async transforms

- Replaced all binaries

- params contains a list of loras, instead of just one

- Added a "spinner" to the `StatelessModeExecute` demo, which spins while waiting for the next token (demonstrating that it's not blocked).

- Removed unused `AddRangeSpan` extension

fix opaque GetState (fixes #176)

make InferenceParams a record so we can use `with`

CPU Feature Detection

MacOS default build now is metal llama.cpp #2901

Removed unused properties of `InferenceParams` & `ModelParams`

better_instruct_antiprompt_checking

Metal depends on this file to execute, and MacOS llama.cpp defaults is now METAL.

See on Mac OS enable Metal by default #2901

`GetState()` fix

* Added the low level bindings to beam search.

Changed `OpenOrCreate` to `Create`