…ost, srctblbrows and strsvr)

It was missing implementations for plots 3 and 4.

preceph: fix code_bias_ix system indexing

rtknavi: fix the track plot expand and shrink for plots 3 and 4

Add BDS to the systems in the default process options

rnx2rtkp: default nav sys to include Galileo and BDS

Common product files are using more that the example 4 lines,
so rework the code to handle this without generating error messages.

It uses the 'flag' to indicated that a clock file is expected and to
handle the format changes in clock files from version 3.04, but could
incorrectly parse the type and system if this expectation was not
met. Rework this to use the flag to bias the search but to fallback to
the alternative if not met.

…and 2 for all constellations if freqs="L1+L2", else use L1/L2 for GLONASS and freq slots 1 and 3 for all other constellations

rtknavi_qt: correct freq change in SNR and sky plots

rtkpost_qt, rtknavi_qt: fix for incorrect target install in CMakeLists.txt

code2bias: guard code being zero, and code_ix not found

readsp3: handle arbitrary comment lines

rinex header: fix clock file system parsing

- Fix field length errors
- Derive IODEs per Beidou ICD

…ndle this case.

…al for SSR and post-processing solutions

- Add reset of rtksvr structure when starting RTKNAVI as a commented out line, uncommenting this line makes RTKNAVI solutions repeatable when postprocessing data from files.

The standard deviations can be supplied at a lower rate than the clock
biases, e.g. 30 sec biases with 5 minute standard deviations seen in
the CODE precise clock bias products.

satposs() was failing when precise clocks were configured and supplied in only
an sp3 file and without broadcast ephemeris. Rework the code to fix this and
a few related issues noted.

pephclk() - rework to search both the precise clocks and the precise ephemeris
data, giving preference to the precise clocks which are typically at a shorter
period and more accurate. It was searching only the precise clocks and failing
to make use of the satellite clock biases in sp3 files in some paths. Return 0
on all failure return paths, so the caller can test the return value for
success or failure, and rework the callers as necessary.

satposs() use the updated pephclk() to work with only satellite clock biases
in sp3 files, and without broadcast ephemeris to give the satellite clock
bias.

pephpos() correct the returned clock variance which was not being scaled by
CLIGHT to return it in units of m^2. It was effectively only applying the
interpolation variance.

Lower the precise clock extrapolation error from 1mm/s to 0.4mm/s. This gives
a midpoint standard deviation of around 6mm for 30s data and 60mm for 5 minute
data which seems more appropriate for current satellite clocks and still
conservative.

satposs: fix to work with precise clocks from sp3 files

readrnxclk: interpolate the standard deviations

… ppp.c to avoid using wrong macro define for state indices

SSR7 types are in common uses, updated from the tentative
allocation. Allows these to be processed with current SSR streams.

Change SSR7 BDS prn offset of 1 to 0 which appears consistent with
current SSR streams, fixing the BDS PRN mapping.

The SSR7 phase bias standard-deviation field do not appear in current
SSR streams, so remove it. Otherwise the phase biases read as garbage.

The RTCM-SSR and IGS-SSR signal mapping has diverged, so split these
into separate tables. Update these tables with current common signal
mappings.

Include the subtype in trace messages in the ssr code paths.

rtcm3 ssr: update types, codes, BDS offset, remove std-dev

rtksrv update_ssr: remove check for ephemeris iode here