Design suggestion for character U+8607 蘇 in TW and HK variant

### Prerequisites

* [x] If you are reporting an issue that affects glyphs for characters for a particular region or regions, did you verify that the characters are within the supported scope of the region or regions? This generally means GB 18030 or Tōngyòng Guīfàn Hànzìbiǎo (通用规范汉字表) for China, Big Five or CNS 11643 Planes 1 & 2 for Taiwan, HKSCS-2016 for Hong Kong, the JIS standards for Japan, and KS X 1001 and KS X 1002 for Korea.
* [x] Did you thoroughly search the [open](https://github.com/adobe-fonts/source-han-serif/issues) and [closed](https://github.com/adobe-fonts/source-han-serif/issues?q=is%3Aissue+is%3Aclosed) issues to avoid reporting a duplicate issue?
* [x] Did you go through the [official font readme file](https://github.com/adobe-fonts/source-han-serif/raw/release/SourceHanSerifReadMe.pdf) to better understand the scope of the project, to include the *Known Issues* section at the very end?

### Description


It would look better if the upper 艹 and the lower 穌 are discrete.

Also I noticed that the bottom of the left part of 艹 is not smooth, maybe that's an issue as well?

I put the other variant alongside just to show how it would look better. Didn't mean it was mapped wrong. I know the 艹 part should be split in TW and HK standard. It should be adjusted to look better.