Add Tokenizer::tokenize_with_location() API

[segoon]

• add TokenWithLocation to track Token location
• implement Tokenizer::tokenize_with_location() -> Result<Vec<TokenWithLocation>, ..>
• refactor line/column tracking to alter current position along with peekable.next() (the old approach with duplicate position tracking was error-prone and contained a bug in multiline word parsing, see the test)
• eliminate String copying for Tokenizer, use &'a str instead
• minor TokenizerError column fixes, now it shows the missing symbol position and not the expected token start location

[coveralls]

Pull Request Test Coverage Report for Build 472821192

• 140 of 147 (95.24%) changed or added relevant lines in 1 file are covered.
• 3 unchanged lines in 1 file lost coverage.
• Overall coverage decreased (-0.05%) to 92.028%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
src/tokenizer.rs	140	147	95.24%

Files with Coverage Reduction	New Missed Lines	%
src/tokenizer.rs	3	91.68%

Totals
Change from base Build 469532238:	-0.05%
Covered Lines:	4883
Relevant Lines:	5306

---

💛 - Coveralls

[Dandandan]

This could be copy as well, as it is just a struct with u64,u64

[Dandandan]

This looks really cool @segoon . Could you fix the conflicts?

[alamb]

Hi @segoon -- sorry for the delay in review. I am going to help out now with this repo and we are working to clear the backlog. Is this PR still something you would like to work on to help contribute?

[alamb]

I am closing what look like stale PRs in this repo; I apologize in advance if this is a mistake -- please feel free to reopen if you want to keep working on this issue.