Fixes subsetting in extract_related_topics and data processing for Namibia

[joachim-gassen]

Thank you for this great package! I use it to pull corona-related search volume for my package {tidycovid19}. This PR tries to fix to issues that I encountered using your package

1. Recently, I had users noticing that gtrends() calls fail at times with the error message

Error in `[<-.data.frame`(`*tmp*`, , timevar, value = "subject") : replacement has 1 row, data has 0

This might be related to #352 but in my use case we do not download search volume for specific categories. I traced the error down to extract_related_topics() and identified an issue in the subsetting of raw_data as end is calculated relative to i and then raw_data is subsetted to i:end which seems to make no sense. This PR fixes the issue for me but I am not sure whether it addresses the issue correctly so discretion is advised.

2. Importing the country code data creates missing values for the ISO2c code of Namibia ("NA") as read.csv() treats "NA" as missing by default. The second commit fixes this. The data commit is redundant if you run process_data.R after applying the commit.

Thank you for looking into this!

Joachim

[PMassicotte]

Thank you, Merged.