[Doc]: fixing typos in multiple files. (#29685)

Signed-off-by: Didier Durand <durand.didier@gmail.com>

vllm/compilation/sequence_parallelism.py

@@ -304,7 +304,7 @@ class SequenceParallelismPass(VllmPatternMatcherPass):
     def __init__(self, config: VllmConfig):
         super().__init__(config)
 
-        # Used to cleanup redundant views created temporarily
+        # Used to clean up redundant views created temporarily
         # to circumvent residual shape change issues
         self.noop_cleanup = NoOpEliminationPass(config)
         self.noop_cleanup.pass_name = f"{self.pass_name}.{self.noop_cleanup.pass_name}"

vllm/model_executor/models/nano_nemotron_vl.py

@@ -1472,7 +1472,7 @@ class NemotronH_Nano_VL_V2(
             return []
 
         # # The result multimodal_embeddings is tuple of tensors, with each
-        # tensor correspoending to a multimodal data item (image or video).
+        # tensor corresponding to a multimodal data item (image or video).
         multimodal_embeddings: tuple[torch.Tensor, ...] = ()
 
         # NOTE: It is important to iterate over the keys in this dictionary

vllm/model_executor/models/stablelm.py

@@ -19,7 +19,7 @@
 # This code is based off the following work:
 # https://huggingface.co/stabilityai/stablelm-3b-4e1t/blob/main/modeling_stablelm_epoch.py
 # https://huggingface.co/stabilityai/stablelm-3b-4e1t/blob/main/config.json
-"""Inference-only StabeLM (https://github.com/Stability-AI/StableLM)
+"""Inference-only StableLM (https://github.com/Stability-AI/StableLM)
 model compatible with HuggingFace weights."""
 
 from collections.abc import Iterable

vllm/utils/gc_utils.py

@@ -17,7 +17,7 @@ class GCDebugConfig:
     """
     Config for GC Debugger.
     - 0: disable GC debugger
-    - 1: enable GC debugger with gc.collect elpased times
+    - 1: enable GC debugger with gc.collect elapsed times
     - '{"top_objects":5}': enable GC debugger with top 5 collected objects
     """
 

vllm/v1/spec_decode/ngram_proposer.py

@@ -269,7 +269,7 @@ def _find_longest_matched_ngram_and_propose_tokens(
                 prev_lps = lps[max_ngram - 1]
             i += 1
         elif prev_lps != 0:
-            # Token mismatch: try the second longest prefix
+            # Token mismatch: try the second-longest prefix
             # among all suffix of tokens[:i],
             # which is the longest prefix of tokens[:prev_lps]
             prev_lps = lps[prev_lps - 1]

vllm/v1/worker/ec_connector_model_runner_mixin.py

@@ -59,7 +59,7 @@ class ECConnectorModelRunnerMixin:
         )
 
     # This context manager must be used within an active forward context.
-    # It encapsulates the entire EC conector lifecycle within execute_model
+    # It encapsulates the entire EC connector lifecycle within execute_model
     @staticmethod
     @contextmanager
     def _get_ec_connector_output(