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/csrc/ktransformers_ext/examples/test_linear.py

@@ -0,0 +1,62 @@
+#!/usr/bin/env python
+# coding=utf-8
+'''
+Description  :  
+Author       : chenht2022
+Date         : 2024-07-25 10:32:05
+Version      : 1.0.0
+LastEditors  : chenht2022 
+LastEditTime : 2024-08-06 10:36:59
+Copyright (c) 2024 by KVCache.AI, All Rights Reserved. 
+'''
+import os, sys
+import time
+sys.path.append(os.path.dirname(__file__) + '/../build')
+import cpuinfer_ext
+import torch
+
+input_size = 16384
+output_size = 5120
+stride = 32
+group_max_len = 1024
+proj_type = 1 # ggml_type::GGML_TYPE_F16
+hidden_type = 1 # ggml_type::GGML_TYPE_F16
+qlen = 30
+layer_num = 10
+CPUInfer = cpuinfer_ext.CPUInfer(48)
+validation_iter = 100
+
+with torch.inference_mode(mode=True):
+    linears = []
+    projs = []
+    for _ in range(layer_num):
+        proj = torch.randn((output_size, input_size), dtype=torch.float16, device = "cuda").to("cpu").contiguous()
+        config = cpuinfer_ext.linear.LinearConfig(input_size, output_size, stride, group_max_len, proj.data_ptr(), proj_type, hidden_type)
+        linear = cpuinfer_ext.linear.Linear(config)
+        projs.append(proj)
+        linears.append(linear)
+
+    # validation
+    for i in range(validation_iter):
+        linear = linears[i % layer_num]
+        input = torch.randn((qlen, input_size), dtype=torch.float16).contiguous()
+        output = torch.empty((qlen, output_size), dtype=torch.float16).contiguous()
+        input = input / 100
+
+        CPUInfer.submit(
+            linear.forward(
+                qlen,
+                input.data_ptr(),
+                output.data_ptr()
+            )
+        )
+        CPUInfer.sync()
+        # print('cpuinfer output', output)
+
+        proj = projs[i%layer_num]
+        t_output = torch.mm(input, proj.t())
+        # print('torch output', t_output)
+
+        diff = torch.mean(torch.abs(output - t_output)) / torch.mean(torch.abs(t_output))
+        print('diff = ', diff)
+        assert(diff < 0.001)