[Misc] Update TokenizerLike interface and move get_cached_tokenizer (#29730)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

.buildkite/test-amd.yaml

@@ -61,8 +61,8 @@ steps:
   - pytest -v -s -m 'not cpu_test' multimodal
   - pytest -v -s utils_
 
-- label: Async Engine, Inputs, Utils, Worker, Config Test (CPU) # 4 mins
-  timeout_in_minutes: 10
+- label: Async Engine, Inputs, Utils, Worker, Config Test (CPU) # 15min
+  timeout_in_minutes: 20
   mirror_hardwares: [amdexperimental, amdproduction]
   agent_pool: mi325_1
   # grade: Blocking
@@ -72,6 +72,7 @@ steps:
   - tests/test_outputs.py
   - tests/multimodal
   - tests/standalone_tests/lazy_imports.py
+  - tests/tokenizers_
   - tests/transformers_utils
   - tests/config
   no_gpu: true
@@ -80,6 +81,7 @@ steps:
   - pytest -v -s test_inputs.py
   - pytest -v -s test_outputs.py
   - pytest -v -s -m 'cpu_test' multimodal
+  - pytest -v -s tokenizers_
   - pytest -v -s transformers_utils
   - pytest -v -s config
 
@@ -308,23 +310,20 @@ steps:
   - pytest -v -s test_regression.py
   working_dir: "/vllm-workspace/tests" # optional
 
-- label: Engine Test # 25min
-  timeout_in_minutes: 40
+- label: Engine Test # 9min
+  timeout_in_minutes: 15
   mirror_hardwares: [amdexperimental, amdproduction]
   agent_pool: mi325_1
   # grade: Blocking
   source_file_dependencies:
   - vllm/
   - tests/engine
-  - tests/tokenizers_
   - tests/test_sequence
   - tests/test_config
   - tests/test_logger
   - tests/test_vllm_port
   commands:
   - pytest -v -s engine test_sequence.py test_config.py test_logger.py test_vllm_port.py
-  # OOM in the CI unless we run this separately
-  - pytest -v -s tokenizers_
 
 - label: V1 Test e2e + engine # 30min
   timeout_in_minutes: 45