using bash expansion expression, combined with mkdir -p (--parents), allows us to create all of the directories at once.

`deps/{avx,avx2,avx512,osx-arm64,osx-x64,cu11.7.1,cu12.1.0,clblast}`
expands to: deps/avx deps/avx2 deps/avx512 deps/osx-arm64 deps/osx-x64 deps/cu11.7.1 deps/cu12.1.0 deps/clblast

I am not on my game today 🤦

GHA doesn't seem to load the value into the enviroment variable first, so the workflow always runs on master.

Checkout specific ref for llamacpp when building native libs

Use llama instead of libllama in `[DllImport]`

This results in windows users not needing to rename the DLL. This allows native llama builds to be dropped in, even on windows.

I also took the time to update the documentation, removing references to renaming the files, since the names now match.

Fixes #463

build(deps): bump System.Text.Json from 8.0.0 to 8.0.1

Bumps [System.Text.Json](https://github.com/dotnet/runtime) from 8.0.0 to 8.0.1.
- [Release notes](https://github.com/dotnet/runtime/releases)
- [Commits](https://github.com/dotnet/runtime/compare/v8.0.0...v8.0.1)

---
updated-dependencies:
- dependency-name: System.Text.Json
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>

kv_cache_instance_methods

Fix Master Build Fail

Swapped `StatelessExecutor` to use `llama_decode`!

build(deps): bump xunit from 2.6.5 to 2.6.6

bump sk & km

- bump semantic kernel to 1.1.0
- bump kernel memory to 0.26

Bumps [xunit](https://github.com/xunit/xunit) from 2.6.5 to 2.6.6.
- [Commits](https://github.com/xunit/xunit/compare/2.6.5...2.6.6)

---
updated-dependencies:
- dependency-name: xunit
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>

LLamaBatch Grow n_seq_max automatically

- Growing more than double, if necessary

- Added `logits_i` argument to `Context.ApplyPenalty`
 - Added a new exception type for `llama_decode` return code

Improved the BatchedDecoding demo

- using less `NativeHandle`
 - Using `StreamingTokenDecoder` instead of obsolete detokenize method

LLamaBatch Automatically Grow Capacity

- Added a `DecodeAsync` overload which runs the work in a task
 - Replaced some `NativeHandle` usage in `BatchedDecoding` with higher level equivalents.
 - Made the `LLamaBatch` grow when token capacity is exceeded, removing the need to manage token capacity externally.

Managed `LLamaBatch`

build(deps): bump Microsoft.AspNetCore.OpenApi from 8.0.0 to 8.0.1

Extra Tokenization Tests

Model Metadata Loading Cleanup

Check Model Path Exists

- Clarified comments on native metadata loading methods

- Checking if model is readable

Bumps [Microsoft.AspNetCore.OpenApi](https://github.com/dotnet/aspnetcore) from 8.0.0 to 8.0.1.
- [Release notes](https://github.com/dotnet/aspnetcore/releases)
- [Changelog](https://github.com/dotnet/aspnetcore/blob/main/docs/ReleasePlanning.md)
- [Commits](https://github.com/dotnet/aspnetcore/compare/v8.0.0...v8.0.1)

---
updated-dependencies:
- dependency-name: Microsoft.AspNetCore.OpenApi
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>

Added a check for EOS token in LLamaStatelessExecutor

Tokenizer Fixes For Issue 430

- Properly displaying `LLamaToken`
 - Removed all tokenisation code in `SafeLLamaContextHandle` - just pass it all through to the `SafeLlamaModelHandle`
 - Improved `SafeLlamaModelHandle` tokenisation:
   - Renting an array, for one less allocation
   - Not using `&tokens[0]` to take a pointer to an array, this is redundant and doesn't work on empty arrays

StreamingTextDecoder Fix & Tests

Update README.md

LLamaToken Struct