[Lora] Support long context lora (#4787)

Currently we need to call rotary embedding kernel for each LoRA, which makes it hard to serve multiple long context length LoRA. Add batched rotary embedding kernel and pipe it through.

It replaces the rotary embedding layer to the one that is aware of multiple cos-sin-cache per scaling factors.

Follow up of https://github.com/vllm-project/vllm/pull/3095/files

format.sh

@@ -112,7 +112,7 @@ mypy vllm/model_executor --config-file pyproject.toml
 
 
 CODESPELL_EXCLUDES=(
-    '--skip' '*docs/source/_build/**'
+    '--skip' '*docs/source/_build/**,./tests/lora/data'
 )
 
 # check spelling of specified files
@@ -133,10 +133,9 @@ spell_check_changed() {
     # `diff-filter=ACM` and $MERGEBASE is to ensure we only lint files that
     # exist on both branches.
     MERGEBASE="$(git merge-base origin/main HEAD)"
-
     if ! git diff --diff-filter=ACM --quiet --exit-code "$MERGEBASE" -- '*.py' '*.pyi' &>/dev/null; then
         git diff --name-only --diff-filter=ACM "$MERGEBASE" -- '*.py' '*.pyi' | xargs \
-             codespell "${CODESPELL_EXCLUDES[@]}"
+            codespell "${CODESPELL_EXCLUDES[@]}"
     fi
 }