Unify `Point`'s and `InfluxDBMapper`'s POJO saving logic

[eranl]

Currently, there are two duplicate and somewhat divergent versions of this logic:

• Point doesn't check that values conform to field types and doesn't constrain allowed value types
• InfluxDBMapper uses a time field rather than @TimeColumn

This change unifies the logic and takes the stricter path, which means that it may break existing code that relies on the current lenient logic.

[codecov-commenter]

Codecov Report

Attention: 6 lines in your changes are missing coverage. Please review.

Comparison is base (d697143) 61.35% compared to head (abbdf75) 60.98%.

:exclamation: Current head abbdf75 differs from pull request most recent head 950708f. Consider uploading reports for the commit 950708f to get more accurate results

Files	Patch %	Lines
src/main/java/org/influxdb/dto/Point.java	78.94%	1 Missing and 3 partials :warning:
...rc/main/java/org/influxdb/impl/InfluxDBMapper.java	75.00%	1 Missing and 1 partial :warning:

:exclamation: Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

```diff @@ Coverage Diff @@ ## master #979 +/- ## ============================================ - Coverage 61.35% 60.98% -0.38% + Complexity 450 435 -15 ============================================ Files 70 70 Lines 2593 2568 -25 Branches 278 276 -2 ============================================ - Hits 1591 1566 -25 Misses 932 932 Partials 70 70 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.