Refactor: restructure repository to focus on kt-kernel and KT-SFT modulesq recon (#1581)

* refactor: move legacy code to archive/ directory

  - Moved ktransformers, csrc, third_party, merge_tensors to archive/
  - Moved build scripts and configurations to archive/
  - Kept kt-kernel, KT-SFT, doc, and README files in root
  - Preserved complete git history for all moved files

* refactor: restructure repository to focus on kt-kernel and KT-SFT modules

* fix README

* fix README

* fix README

* fix README

* docs: add performance benchmarks to kt-kernel section

Add comprehensive performance data for kt-kernel to match KT-SFT's presentation:
- AMX kernel optimization: 21.3 TFLOPS (3.9× faster than PyTorch)
- Prefill phase: up to 20× speedup vs baseline
- Decode phase: up to 4× speedup
- NUMA optimization: up to 63% throughput improvement
- Multi-GPU (8×L20): 227.85 tokens/s total throughput with DeepSeek-R1 FP8

Source: https://lmsys.org/blog/2025-10-22-KTransformers/

This provides users with concrete performance metrics for both core modules,
making it easier to understand the capabilities of each component.

* refactor: improve kt-kernel performance data with specific hardware and models

Replace generic performance descriptions with concrete benchmarks:
- Specify exact hardware: 8×L20 GPU + Xeon Gold 6454S, Single/Dual-socket Xeon + AMX
- Include specific models: DeepSeek-R1-0528 (FP8), DeepSeek-V3 (671B)
- Show detailed metrics: total throughput, output throughput, concurrency details
- Match KT-SFT presentation style for consistency

This provides users with actionable performance data they can use to evaluate
hardware requirements and expected performance for their use cases.

* fix README

* docs: clean up performance table and improve formatting

* add pic for README

* refactor: simplify .gitmodules and backup legacy submodules

- Remove 7 legacy submodules from root .gitmodules (archive/third_party/*)
- Keep only 2 active submodules for kt-kernel (llama.cpp, pybind11)
- Backup complete .gitmodules to archive/.gitmodules
- Add documentation in archive/README.md for researchers who need legacy submodules

This reduces initial clone size by ~500MB and avoids downloading unused dependencies.

* refactor: move doc/ back to root directory

Keep documentation in root for easier access and maintenance.

* refactor: consolidate all images to doc/assets/

- Move kt-kernel/assets/heterogeneous_computing.png to doc/assets/
- Remove KT-SFT/assets/ (images already in doc/assets/)
- Update KT-SFT/README.md image references to ../doc/assets/
- Eliminates ~7.9MB image duplication
- Centralizes all documentation assets in one location

* fix pic path for README

archive/ktransformers/tests/dequant_gpu.py

@@ -0,0 +1,58 @@
+import os 
+# os.environ["CUDA_VISIBLE_DEVICES"]="1,2"
+# add path
+import sys
+current_path = os.path.abspath(os.path.dirname(__file__))
+sys.path.append(current_path+"/../..")
+import numpy as np
+# from ktransformers.operators.linear import KTransformersLinear, KLinearMarlin
+# from ktransformers.operators.experts import KTransformersExperts, KExpertsTorch
+from ktransformers.util.custom_gguf import GGUFLoader
+import torch
+import KTransformersOps
+torch.set_default_dtype(torch.bfloat16)
+import time
+from transformers import (
+    AutoConfig,
+)
+import os
+# CUDA_LAUNCH_BLOCKING=1
+os.environ["CUDA_LAUNCH_BLOCKING"]="1"
+
+gguf_config = GGUFLoader("/data/Qwen2-57B-A14B-Instruct-GGUF/q4_k_m")
+model_name = "/data/Qwen2-57B-A14B-Instruct"
+
+# Q4k
+key = "blk.1."
+target = "attn_q.weight"
+
+t1 = time.time()
+q_weight_cpu = gguf_config.load_gguf_tensor(key+target, "cpu")
+# q_weight_cpu = torch.from_numpy(q_weight_cpu)
+
+t2 = time.time()
+q_weight_gpu = gguf_config.load_gguf_tensor(key+target, "cuda:0")
+t3 = time.time()
+print()
+allclose = torch.allclose(q_weight_cpu, q_weight_gpu.cpu(), atol=1e-6)
+print(f"Q4k {key+target}")
+print("load gguf tensor from cpu cost: ", t2-t1)
+print("load gguf tensor from gpu cost: ", t3-t2)
+print("allclose: ", allclose)
+
+
+# Q6k
+key = "blk.0."
+target = "ffn_down_exps.weight"
+
+t1 = time.time()
+q_weight_cpu = gguf_config.load_gguf_tensor(key+target, "cpu")
+t2 = time.time()
+q_weight_gpu = gguf_config.load_gguf_tensor(key+target, "cuda:0")
+t3 = time.time()
+print()
+allclose = torch.allclose(q_weight_cpu, q_weight_gpu.cpu().to(torch.float32), atol=1e-6)
+print(f"Q6k {key+target}")
+print("load gguf tensor from cpu cost: ", t2-t1)
+print("load gguf tensor from gpu cost: ", t3-t2)
+print("allclose: ", allclose)