对其中一个问题，补充up写过的知乎实例。

* update faq.md about ncnn-produce-wrong-result

* update faq.md  about print Mat content and visualize Mat content

* update faq.md about efficient roi resize rotate

* update faq.md about convert to fp16

* update faq.md about use ncnn with pytorch or onnx

* update faq.md about add the url of netron

* update faq.md about add the url of convertmodel

* update faq.md about the url of darknet2ncnn

* update faq.md about tensorflow 2.x-mlir

* update faq.md about keras2ncnn

* update faq.md about Post Traning Quantization tools

* update faq.md about caffe2ncnn

* update faq.md about mxnet2ncnn

* update faq.md about bf16

Co-authored-by: runrunrun1994 <runrunrun1994@163.com>

* update faq.md about ncnn-produce-wrong-result

* update faq.md  about print Mat content and visualize Mat content

* update faq.md about efficient roi resize rotate

* update faq.md about convert to fp16

* update faq.md about use ncnn with pytorch or onnx

* update faq.md about add the url of netron

* update faq.md about add the url of convertmodel

Co-authored-by: runrunrun1994 <runrunrun1994@163.com>

* doc: add build-mlir2ncnn with commit sha and steps

* add mlir git log detail

* fix typos in docs

* still fix some typos in docs

Co-authored-by: wq.chu <wq.chu@tianrang-inc.com>

* Add [ MacOS ->  apple-cpu/apple-gpu] into supported platform matrix
 * Add [ mac ->  apple] into 'device support vulkan' list

Co-authored-by: meow <826150766@qq.com>

* add convolutiondepthwise/crop/sigmoid/tanh in operators.md

* sort by alphabet

* delete impl_type

* cannot delete impl_type

* cannot delete impl_type

Co-authored-by: yaobin.li <yaobin.li@rokid.com>

* update operators.md, add binaryop/Operation type

* fix with_scalar

* Update operators.md

* Update operators.md

* Add LSTM and softmax to operators.md

* Fix lowercase and support_inplace

prvious -> previous

* add missing word

* fix wrong-spelled apt package names

crash when freeing a ncnn dynamic library

* Improve build instructions

* Merge Linux stuff into one section.
* Provide better info on dependencies
* Provide easier to use instruction for Debian / Ubuntu for native and cross-compile builds
* Capitalize Sentence and put proper punctation.
* Use backticks where appropiate
* Move Hisilicon instruction together with Arm crosscompiling
* Add information how to verify the Vulkan installation on Linux
* Change how to launch examples so it is easier (no copying needed or external resources)
* Add info about benchmark and how to test it on GPU or CPU
* Use `$(nproc)` where appropriate.
* Put some paths in quotes, in case they contain spaces or other weird characters (coma, semicolon, colon, etc).
* Remove some extremely obvious pieces that every developer usually knows.
* Use newer Vulkan SDK in instructions for Linux
* Get rid of `<ncnn-root-dir>`, and just use `ncnn`. It is what git clone does. And any developer will know how to change it if they choose to do it differently.

* Update README.md

* Update how-to-build.md

Co-authored-by: nihui <shuizhuyuanluo@126.com>