fix a panic causing a crash when clangd sends workspaceSemanticTokensRefresh

[santtuniskanen]

Please check if the PR fulfills these requirements

• The PR has no duplicates (please search among the Pull Requests
  before creating one)
• The PR follows our contributing guidelines
• Tests for the changes have been added (for bug fixes / features)

• What kind of change does this PR introduce?
  adds a client handler for workspace/semanticTokens/refresh in lsp_client_clangd.go.

• What is the current behavior?
  When clangd sends workspace/semanticTokens/refresh, it causes a panic. This causes the language server to crash on every restart. A workaround has been to disable semantic tokens in your editor, but this fix should prevent the need to do so.

21:11:04.196586 IDE LS <-- Clangd REQU workspace/semanticTokens/refresh 1
21:11:04.196684 Panic: unimplemented request: workspace/semanticTokens/refresh

• What is the new behavior?
  refresh request comes in and goes out, without causing a panic and crashing the language server.

21:19:11.867890 IDE LS <-- Clangd REQU workspace/semanticTokens/refresh 1
21:19:11.867912 IDE LS --> Clangd RESP workspace/semanticTokens/refresh 1

• Other information:
  depends on bugst/go-lsp#10 to be approved and merged first. The module version should then be bumped up before merge.

---

Fixes #125 Fixes #155

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.

---

Santtu Niskanen seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}