[TRTLLM-4501][feat] AutoTuner tuning config refactor and add tuning for kernel configs.

[hyukn]

The motivation for this PR is #4872, in which AutoTuner is applied to FP8 batched GEMM op with tile_size and epilog_tile_m to be in the argument list. Generally, there are two possible implementations.

• Encoding different configs into a list of numeric tactic IDs starting from 0. This will be implemented inside kernels and used through get_valid_tactics.
• Define each config separately and let AutoTuner iterate over the combinations. This is more readable and flexible. Users can use each part of the config directly. There is no encoding-decoding process.

We choose the second method: Adding a config entry in the tuning config to define the valid candidates for each part of the config.

• AutoTuner will loop over a search grid generated from the config combinations.
• Each config will be tuned along with the specific input profile.
• The best config will be recorded in the cache value (instead of the cache key). And it will be recovered and used in the tunable runner forward.

Other enhancements:

• Use the decorator to make the tuning config definition more natural and efficient. This is an independent enhancement.
• Allow the user to not specify the gen_tuning_buckets or the map_to_tuning_buckets function.
• Code refactoring.

[hyukn]

/bot run

[tensorrt-cicd]

PR_Github #10964 [ run ] triggered by Bot

[hyukn]

/bot run

[tensorrt-cicd]

PR_Github #10979 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #10964 [ run ] completed with state ABORTED

[tensorrt-cicd]

PR_Github #10979 [ run ] completed with state FAILURE
/LLM/main/L0_MergeRequest_PR pipeline #8109 completed with status: 'FAILURE'

[hyukn]

/bot run

[tensorrt-cicd]

PR_Github #11098 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #11098 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8205 completed with status: 'FAILURE'

[hyukn]

/bot run

[tensorrt-cicd]

PR_Github #11109 [ run ] triggered by Bot

[hyukn]

/bot run

[tensorrt-cicd]

PR_Github #11111 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #11109 [ run ] completed with state ABORTED

[tensorrt-cicd]

PR_Github #11111 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8217 completed with status: 'FAILURE'

[hyukn]

/bot run

[tensorrt-cicd]

PR_Github #11168 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #11168 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8261 completed with status: 'FAILURE'

[hyukn]

/bot run

[tensorrt-cicd]

PR_Github #11203 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #11203 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8287 completed with status: 'FAILURE'

[hyukn]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #11234 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #11234 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8309 completed with status: 'FAILURE'

[hyukn]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #11414 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #11414 [ run ] completed with state FAILURE
/LLM/main/L0_MergeRequest_PR pipeline #8442 completed with status: 'FAILURE'

[hyukn]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #11483 [ run ] triggered by Bot

[tensorrt-cicd]

PR_Github #11483 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #8496 completed with status: 'SUCCESS'