sqanti_qa report fails if there are no isoforms in a category

[diekhans]

If a category has no isoforms it, errors like the following occur:

       Quitting from lines 385-393 [p4.s1_p4.s2_p4.s3] (SQANTI3_report.Rmd)
         Error in `ans[ypos] <- rep(yes, length.out = len)[ypos]`:
         ! replacement has length zero
         Backtrace:
           1. ggplotly(p4.s3) %>% layout(showlegend = T, boxmode = "group")
           4. plotly:::ggplotly.ggplot(p4.s3)
           5. plotly::gg2list(...)
           6. plotly (local) ggplotly_build(p)
           7. layout$setup_panel_params()
              ...
          15. ggplot2:::expand_limits_scale(scale, expansion, limits, coord_limits = coord_limits)
          16. ggplot2:::expand_limits_discrete(...)
          17. ggplot2:::expand_limits_discrete_trans(...)
          18. ggplot2:::expand_limits_continuous_trans(...)
          19. base::ifelse(is.na(coord_limits), limits, coord_limits)
         Execution halted

This occurred by filtering all of the isoforms for novel gene categories and re-running sqanti on clean up set of isoforms. I will be submitted a pull request that helped work around this.

[almart7]

Dear @diekhans thank you so much for your input! We will review it as soon as possible.

[aarzalluz]

@diekhans merged PR #222, fix will be available in next SQANTI3 release. Thanks!

Issue #219 will remain open until we can further enhance the report plots to prevent errors when categories have no isoforms.